AppWorx 9.4 User Guide
AppWorx 9.4 User Guide
Copyright © 2005-2019 Broadcom. All Rights Reserved. The term 'Broadcom' refers to Broadcom Inc. and/or its subsidiaries.
Contents
Release Information................................................................................................................... 11
Release Notes................................................................................................................... 11
Third-Party Licenses and Notices..................................................................................... 14
Installation Guide........................................................................................................................76
About This Guide.............................................................................................................. 77
Applications Manager Installation for UNIX and Windows................................................77
Creating the UNIX Account (UNIX Only)................................................................. 78
Assigning Rights and Permissions to the Windows User (Windows Only).............. 79
Creating the Database Account............................................................................... 80
System Information Required for the Installation.....................................................82
Running the Installation Script................................................................................. 84
Starting and Stopping the Applications Manager Processes in UNIX......................85
Starting and Stopping the Applications Manager Processes in Windows................ 86
Opening the Applications Manager Client and Logging In.......................................87
Verifying the Installation........................................................................................... 90
Applications Manager Installation—Advanced Topics.......................................................93
Configuring the Applications Manager Client...........................................................93
Function of the Operating System User.................................................................. 93
Configuring Applications Manager for Oracle RAC................................................. 95
Installing Multiple Automation Engines on One Host...............................................97
Configuration for Machines with Multiple IP Addresses...........................................98
Automation Engine, Agent, and Client Firewall Connections.................................100
Overview of Firewall Settings.................................................................................102
Configuring Agents to Validate Multiple Automation Engine Configurations.......... 102
Using Custom SSL Certificates for Connection Authentication..............................103
Uninstalling Applications Manager......................................................................... 104
Installing a Second RMI Server for Failover.......................................................... 105
Installing a UNIX Remote Agent.....................................................................................107
Defining the UNIX Remote Agent in Applications Manager...................................108
Installing the Remote Agent...................................................................................109
Installing a Windows Remote Agent............................................................................... 110
Defining the Windows Remote Agent in Applications Manager.............................111
Creating the Windows User................................................................................... 113
Running the Installation Program...........................................................................114
Upgrading Applications Manager.................................................................................... 115
Upgrading or updating Tomcat...............................................................................115
Upgrading an Automation Engine and Local Agent...............................................116
Upgrading Remote Agents.....................................................................................118
Verifying the Upgrade.............................................................................................119
Loading Rapid Automation Component .jar Files into Applications Manager................. 120
1 Release Information
Release highlights for Applications Manager 9.x
Warning: Broadcom highly recommends that you back up your Oracle data and Applications Manager file
system before upgrading due to table changes in this release. There is no rollback procedure for Oracle
data and the upgrade process does not back up your data or file system.
New Features Only in Major and Minor Versions of All Automic Software
Aways install the latest Service Pack or Hotfix. Both contain valuable corrections and bug fixes between Major and
Minor Releases, where new features and enhancements are introduced.
The software version information consists of the following identifiers, as shown in this example:
12.2.1 HF 1
• Major Release: 12
This is the main version of a software release. It is identified by the first segment of the entire version number.
• Minor Release: 2
This includes new features, modifications, and corrections that may contain major changes such as database
modifications. It is identified by the second segment of the entire version number.
Major and Minor releases for Automation Engine are feature releases supplied at 9-month intervals.
• Service Pack: 1
This is a patch for a release and contains corrections for errors. New features or modifications are not included.
Service packs are identified by the third segment of the entire version number.
• Hotfix: HF1
This is a minor sub-release to remove malfunctions and defects. Hotfixes are indicated by an HF number after
the version number.
Service Packs and Hotfixes are maintenance releases supplied at 4-month intervals.
Getting the Latest Information
Documentation, release notes, and other information is often updated after software is released. The table below
shows where to find the most recent information for Automic software releases.
Release Notes
Applications Manager V9.4.1
Applications Manager 9.4.1
What's New?
Applications Manager 9.4.1 addresses major vulnerabilities reported with Log4j libraries. It also includes several
bug fixes.
Applications Manager V9.4.0
What's New?
• Applications Manager 9.4.0 comes with improved disaster recovery management. For more information, refer to
Disaster Recovery.
• Apache Tomcat is now the default HTTP server. For more information refer to Upgrading or updating Tomcat.
• Starting from Applications Manager 9.4.0, Perl is no longer being shipped with the product image for UNIX-
based operating systems.
• New fields are introduced for Banner Output.
• Three new grants are added for the Banner database SYS user. For more information refer to, Issuing Grants to
Create the BANWORX Banner User.
• The product help will now be available online only.
Applications Manager V9.3.5
What's New?
• SMTP Authentication: You can now configure SMTP authentication for the master agent. For more information,
refer to Configure SMTP Authentication.
• Custom SSL certificates for Connection Authentication: Improvements have been made for use SSL certificates.
For more information, refer to Using Custom SSL Certificates for Connection Authentication.
• Notifications: You can now attach a PDF to the notification in the portrait or landscape format with size limit
option. For more information, refer to Defining Notifications.
Applications Manager V9.3.4
What's New?
• Broadcom may request you for the telemetry report as and when required. For more information, refer to
Applications Manager Telemetry data.
• You can now add a component with a right-click to an existing job or process flow and also retain the
component's configuration. For more information, refer to Add components and retain the existing configuration.
• You can now have different colors for Applications Manager main window for different deployments of
Applications Manager. For example, you may want to have a Brown color for the development environment and
Green color for the production environment. For more information, refer to Configure Applications Manager Main
Window Color.
• Applications Manager 9.3.4 now supports RedHat Enterprise Linux 8.
Applications Manager v9.3
What's New
Oracle 18c Now Supported
In Applications Manager 9.3, Oracle 18c databases are supported. For special instructions, see Creating the
Database Account in the Installation Guide.
Specifying a Unique Location for Multiple Keystore Files in v9.3.1
Starting with v9.3.1, you can specify a unique location for multiple keystore files. To do this, you create a C:\Users
\<user name>\AppWorx\<master name> folder for each master in the connections.properties file on each
user's client machine where <user name> is the actual user's name and <master name> is the name of the
master. Then put copies of the user_keystore and user_keystore_config files for each master the sub-directory
for that master.
RunClient.jar Error Messages Now in the startClient.log file in v9.3.1
When starting the local client by double clicking on RunClient.jar (or a shortcut to it) in version 9.3.1 or above, if it
fails to start the error will now be in the startClient.log file.
Changes in Behavior
Executable .jar File for the Client Must Be Downloaded and Configured
You must now download the Applications Manager client from the URL set during installation. To get this URL,
see your Applications Manager administrator. Before logging in the first time, you need to configure the client for
Applications Manager 9.4.1
the Automation Engine(s) you will be logging into. Once configured, you open an executable .jar file. For more
information, see Opening the Applications Manager Client and Logging In in the User Guide.
Custom SSL Certificates for Connection Authentication Now Required
You must use a custom SSL certificate for connection authentication by creating user_keystore and
user_keystore_config files on the Automation Engine and client machines. For more information, see Using
Custom SSL Certificates for Connection Authentication in the Installation Guide.
Keyfiles No Longer Required
Starting with Applications Manager 9.3, license checking with keyfiles is no longer required for Automation Engines
or agents.
integrated with the specific Automation Engine system. Therefore, in systems that require RMI server failover, it is
recommended that all Jobs be setup to run on Remote Agents.
Platform and Oracle Support
Version 9 is the first Applications Manager release that supports Oracle 12c databases. When using Oracle 12c
database and you have an Oracle Enterprise Edition multi-tenant system or Oracle Standard Edition single tenant
system, you must install Applications Manager into a pluggable database (PDB) within a container. Your DBA is
responsible for creating the pluggable database.
Enabling/Disabling RMI Server Debug with the RMI_DEBUG Job
You can now enable or disable RMI server debug by running the RMI_DEBUG Job. Like all Applications Manager
Jobs, you can request and submit RMI_DEBUG to run it ad hoc, or schedule it. Activate or deactivate RMI server
debug by setting the Enable prompt to Yes or No. The RMI_DEBUG Job is included in the RMI_DEBUG import file.
Desupported Features
The following features are desupported for Applications Manager v9:
• The Dashboard
• AppMaster
• Remote Method Extension Agents
• VMS Agents
• OS400 Agents
• ZOS (mainframe bridge) Agents
The supported platforms for the Automation Engine and its Agents change between releases. For a list of
supported platforms for the Automation Engine and Agents in this release, see the Matrix below.
Known Issue: You Cannot Retrieve a Local WSDL File for a SOAP Login Object
You cannot retrieve a local WSDL file for a SOAP Login object using the Web Service Agent. This will be fixed in
Applications Manager v9.0 Service Pack 1 and a patch is available from Broadcom Support.
• Jobs: are the basic building blocks in Applications Manager. For each program you want to run (such as FTP,
application, or database load), you must create a Job. A Job contains all the information required to execute
a program and handle its output. Jobs are run both individually and as components of Applications Manager
Process Flows. Furthermore, a Job can be a component of as many Process Flows as you wish. If you change
a Job definition, the change is applied to every Process Flow that includes it. For more information, see Defining
Jobs.
• Libraries: define the path for the program that a Job runs. For more information, see Defining Libraries.
• Message Templates: specify text to include in Notification output files. They allow the use of Substitution
Variables and Replacement Values for variable text. For more information, see Defining Message Templates.
• Notifications: send dynamic messages that are based upon on a task's status through email or any other
defined Output Device. For more information, see Defining Notifications.
• Output Scans: scan output for text strings that indicate if a task has failed or succeeded. For more information,
see Defining Output Scans.
• Process Flows: are containers that include one or more components (Jobs and other Process Flows), general
scheduling information for the Process Flow, specific eligibility for each of its components, and conditions that
must be met for each component to run. For more information, see Defining Process Flows.
• Program Types: define how programs accept input and handle output. For more information, see Defining
Program Types.
• Reports: view Reports for data in the Applications Manager database. Developers and administrators can
create custom SQL Reports in the Applications Manager client. For more information, see Steps for Creating
Reports.
• Substitution Variables: store values that can be used in Jobs and Process Flows. Applications Manager lets
you use Substitution Variables, such as #today, in prompts and execution conditions assigned to Jobs. For more
information, see Defining Static Substitution Variables and Defining Dynamic Substitution Variables.
Operations
• Queues: control the flow of tasks. All tasks must pass through an Applications Manager Queue to be executed.
For more information, see Defining Queues.
• Thread Schedules: are assigned to one or more Queues or Agents. Thread Schedules define the number of
concurrent tasks that can run through a Queue or Agent at different times of the day. For more information, see
Defining Thread Schedules.
Automation Engine
The Automation Engine is the execution logic or brain of Applications Manager. It monitors Job and Process Flow
schedules and, at the appropriate time, sends them to the designated local or Remote Agent for execution. The
Automation Engine communicates with the Applications Manager database, where all object definitions are stored.
Local Agent
At a minimum, a basic Applications Manager setup will include a single Automation Engine and its Local Agent.
The Local Agent runs programs or executes scripts on the host machine where the Applications Manager
Automation Engine is installed. It receives commands from the Automation Engine.
Applications Manager Database
Applications Manager uses object definitions stored in an Oracle database to give it an advantage over all other
schedulers. When you create a Job, Process Flow, or other Applications Manager object, the definition is stored
in the Applications Manager Oracle relational database. You can then use the objects to build more complex
objects. For example, you might define a Database Login, then assign that login to many different Jobs. If the
login changes, you change it in one place, and Applications Manager uses the new definition everywhere it is
referenced. This object-oriented approach makes it easy to update and maintain the Applications Manager system.
Client
You access Applications Manager through a Java client. The client machine is the PC used to access the
Applications Manager graphical user interface. This means you can access Applications Manager from a PC or
workstation that has a Web browser. Client software provides access to all Applications Manager functions and
features. The clients communicate directly with the Applications Manager Automation Engine. You can have any
number of clients.
Remote Agent
Applications Manager 9.4.1
In larger systems, you may have one or more Applications Manager Remote Agents on other servers. A Remote
Agent must be installed on each machine where tasks are executed. Any number of Remote Agents can report to a
single Automation Engine. Remote Agents may run on UNIX and Windows platforms.
The Automation Engine will schedule and control task execution on all the Agents assigned to it. The Agent
monitors tasks until they complete.
Medium System
If you are installing Applications Manager in a medium sized system, you will most likely add several Applications
Manager Remote Agents to execute programs on other machines. This configuration is shown below.
Applications Manager 9.4.1
Large System
In a large system, you may choose to install two or more Applications Manager Automation Engines. This
configuration is shown below.
Applications Manager 9.4.1
An Applications Manager Automation Engine can process hundreds of thousands of tasks a day, so the choice
to install more than one Automation Engine is usually not because of load. A more likely reason is that you
are running Applications Manager in different data processing centers and do not want to depend on network
connections between the Automation Engine and Remote Agents.
Reports
To help you review the objects that have been created, Applications Manager includes many object Reports. The
Reports list all objects of each type, and in some cases, additional information about the objects.
• Agents: are instances of Applications Manager. An Agent is installed on each machine where tasks are
executed. An Agent can be an Automation Engine's Local Agent, a Remote Agent, or an application-specific
Agent such as Oracle Applications or PeopleSoft.
• Logins: allow operators and programmers to run programs that access a database or host without having to
know the login and password.
• Output Devices: define any Output Device including printers, faxes, and email. You can print to a single Output
Device, or, through the use of distribution lists, multiple Output Devices.
• Output Groups: define organizational classes of Output Devices. When you define Applications Manager
Output Devices, you assign them to one or more Output Groups.
• Output Interfaces: interface between Applications Manager and an Output Device.
• User Authorities: control User access to Applications Manager windows and can give Users add privileges for
objects. You assign User Authorities to your User Groups.
• User Groups: control access to all areas of Applications Manager. In a traditional system, you create groups of
Users, Output Devices, and Applications. User Groups can contain any combination of objects, and objects can
be assigned to any number of User Groups.
• User Options: are the same options assigned to Users, but as objects, they are assigned to a User Group.
When User Options are assigned to a User Group, all Users in that User Group will have that User Option set to
true for them. You cannot add, edit, or delete User Option objects like other Applications Manager objects. You
can only assign them to User Groups.
• Users: control access to Applications Manager. You can assign names, access permissions, User Options, and
User Groups to Users.
Development
• Applications: logically group Jobs and Process Flows. Used to filter lists of Jobs and Process Flows.
• Calendars: define groups of days, such as holidays, that you can use for schedules. Schedules can run on, or
skip, the days in a Calendar.
• Data Types: specify the format for data passed to programs. They can include SQL statements that validate
responses and allow you to pick from lists.
• Environment Variables: store values you define for one or more variables as a single Applications Manager
object.
• Jobs: are the basic building blocks in Applications Manager. For each program you want to run (such as FTP,
application, or database load), you must create a Job. A Job contains all the information required to execute
a program and handle its output. Jobs are run both individually and as components of Applications Manager
Process Flows. Furthermore, a Job can be a component of as many Process Flows as you wish. If you change
a Job definition, the change is applied to every Process Flow that includes it.
• Libraries: define the path for the program that a Job runs.
• Message Templates: specify text to include in Notification output files. They allow the use of Substitution
Variables and Replacement Values for variable text.
• Notifications: send dynamic messages that are based upon on a task's status through email or any other
defined Output Device.
• Output Scans: scan output for text strings that indicate if a task has failed or succeeded.
• Process Flows: are containers that include one or more components (Jobs and other Process Flows), general
scheduling information for the Process Flow, specific eligibility for each of its components, and conditions that
must be met for each component to run.
• Program Types: define how programs accept input and handle output.
• Reports: view Reports for data in the Applications Manager database. Developers and administrators can
create custom SQL Reports in the Applications Manager client.
• Substitution Variables: store values that can be used in Jobs and Process Flows. Applications Manager lets
you use Substitution Variables, such as #today, in prompts and execution conditions assigned to Jobs.
Operations
• Queues: control the flow of tasks. All tasks must pass through an Applications Manager Queue to be executed.
• Thread Schedules: are assigned to one or more Queues or Agents. Thread Schedules define the number of
concurrent tasks that can run through a Queue or Agent at different times of the day.
When you log in to Applications Manager, you are taken to the desktop shown below. From the desktop, you can
access all Applications Manager features and functions.
Selector windows are used to manage Applications Manager objects. They are displayed when you select an icon
from the toolbar or an item from the Object Admin menu. There is a selector window for each type of Applications
Manager object. From a selector window, you can add, edit, and delete objects. In the image below, the New button
is selected on the Queues Selector window and a new Queue is being defined.
Administration
If you are a Applications Manager administrator, you will be responsible for ensuring the system is up and running,
and that users can access the system. How much more you are responsible for will depend on the structure of your
IT shop. At the very least, you will probably be expected to:
• Ensure that the Applications Manager Automation Engine, Agents, RMI server, and Web server are up and
running.
• Provide Applications Manager User IDs and Logins to developers, production analysts, operations personnel,
network administrators and database administrators.
• Set the Automation Engine options.
You may also be responsible for one or more of the following:
• Defining Host and Database Logins
• Defining Output Devices
• Managing retention of Applications Manager records and report output
• Migrating Applications Manager objects between development, test, and production instances
This diagram shows a more likely scenario where the Applications Manager Automation Engine, Local Agent, RMI
server, and Web server are all on one machine, and the Applications Manager database is on another machine.
Applications Manager 9.4.1
The login allows a User to enter the Applications Manager client. Once the User is in the client, access to objects
and to different areas of the product, often referred to as managers, is controlled by User Groups.
User Groups Are Important
User Groups control access to the Applications Manager and its objects. When you create a User, you assign one
or more User Groups to the User. You can think of User Groups as containers. After creating a User Group, you
add objects and Users to the User Group. Users have access to all the objects in the container.
It's important for you to give some thought to the User Groups that will best serve the security needs of your
organization. For example, if you are a very small shop, you may need only a few User Groups such as
administrator, developer, and operations. If you are a large shop, you may want to use a more extensive set of User
Groups such as: administrator, financial applications developer, sales inventory application developer, customer
relations application developer, network administrator, database administrator, operator, and production analyst.
Each User Group would give access to different parts of Applications Manager and to a different set of objects.
The top row in the image below illustrates three typical User Groups: programmer, operator, and end-user. The
image also shows two additional programmer User Groups: edit and non-edit. The two User Groups make it
possible to give programmers read-only access to some objects, and edit access to other objects.
The programmers would be assigned to both User Groups. For example, you might give programmers read-only
access to:
• Objects that ship with the product (such as system Jobs and Process Flows).
• Objects that might be created by an Applications Manager administrator (such as Output Devices).
• Objects that might be created by a database administrator (such as Logins).
Applications Manager 9.4.1
• Certain objects such as Queues (to give access to the Explorer window).
On the other hand, you would give programmers edit access to the Jobs and Process Flows that they create.
By having the two User Groups, you have the flexibility to give a group of Users the access they require to
accomplish their Job. You would most likely want an edit and non-edit User Group for operators as well. End-users
may only require a single non-edit User Group because they would not be creating objects.
DBA User Group
When you first install Applications Manager, there is one User Group called "DBA" that is assigned to the default
User. The "DBA" User Group gives the default User access to all functions and objects in Applications Manager.
By logging in as the default User (see the Installation Guide for the default user name and password), you can then
define additional Users.
You can control the Logins that are displayed in the list box by using Applications Manager User Groups. Only the
Logins that are included in a User Group assigned to the developer will be displayed. Security is maintained, and
the developer's task is made easier.
Encrypted Passwords
In both Host and Database Logins, the Login names are displayed as plain text, but the passwords are encrypted.
The only person who knows the password is the person who entered it in Applications Manager. This ensures a
high level of security.
Reduced Maintenance
By defining the Database and Host Logins as objects in Applications Manager, you can update a Login in one place
and have that change take effect everywhere the Login is used. This greatly reduces your maintenance time.
Managing Output
Applications Manager can capture output from any task it runs, send the output to any type of Output Device, and
make the output available for viewing online through the Applications Manager Java client.
Applications Manager 9.4.1
Applications Manager includes a built-in output distribution function that can send output to almost any type of
Output Device including traditional printers, fax machines, plotters, and email. But Applications Manager also can
make the output available for online viewing directly from the Applications Manager client.
file=$1
Applications Manager 9.4.1
shift
eval "lp $* $file"
exit $?
As an Applications Manager administrator, you may define Output Devices in Applications Manager, or you may
provide access to a network administrator to define the device.
Online Viewing
When Applications Manager runs a task, such as an Oracle Applications or PeopleSoft task, it builds the
appropriate commands using an Applications Manager Program Type interface script. Also incorporated in the
script is the ability to register the output from the task and make it available instantly for online viewing. This
capability is an integral part of Applications Manager.
When determining how to make the best use of Applications Manager output management, you should examine
this ability to view output online and how it might affect how you currently distribute output.
Applications Manager interface scripts are available for the most popular enterprise applications, and Broadcom
Global Services Offerings services can create custom scripts for almost any application.
Archiving Records
The SYSTEM Process Flow purges output files and History records, but it does not archive the files or records.
If you wish to perform archive functions, consider building an archive Process Flow that executes the functions.
For example you could create a Process Flow that prints a report of the History records, moves the records to
an archival database, and moves the output files to an archival directory. This archival Process Flow could be
scheduled to run before the SYSTEM process flow.
You should never modify the SYSTEM Process Flow because it is reinstalled every time you upgrade or reinstall
Applications Manager. If you want to modify the SYSTEM Process Flow, copy it to another Process Flow, then
modify the copied Process Flow and run it instead of the SYSTEM process flow.
Moving Applications Manager objects from one instance to another is a simple three-step process:
1. Export the objects from the source machine.
2. Move the export file to the target machine.
3. Import the objects to the target machine.
Exporting Objects
To export objects, use the Exports window and build an export list. The list can include any number of objects. You
can export all Applications Manager objects except Agents, Agent Groups, Logins, and Users. When you select
an object, Applications Manager automatically identifies any supporting objects. For example, if you include a Job,
Applications Manager identifies any Output Devices assigned to the Job and gives you the opportunity to add the
Output Devices to the export list. You can save the export list and reuse it the next time you want to export objects.
After building the export list, you run the export program. The program generates the export file.
Moving the Export File
After generating the export file, you transfer the file from the source machine to the target machine.
Importing Objects
On the target machine, use the Imports window to import the objects from the export file. When you load the file
into the Imports window, support objects such as Agents, Agent Groups, Logins, and Users that could not be
included in the file will be flagged. You can then map these objects to corresponding objects on the target machine.
Applications Manager 9.4.1
For example, suppose you are exporting from a development machine to a test machine. An exported Job was
defined to run on the DEV Agent. On the test machine, the corresponding Agent is TEST. You simply map the DEV
Agent to the TEST Agent.
When you have completed mapping the objects, you can save the map for repeated use, ensuring consistency.
To complete the process, you run the import. The objects are added to the target instance.
Development
As a developer, you will create objects used to define Jobs and run tasks. When we talk about development in
Applications Manager, we are referring to the creation of Jobs to run tasks, and the addition of Jobs to Process
Flows. If you are a developer, you will most likely:
• Create Jobs to run programs and scripts.
• Create Process Flows to run a series of Jobs.
• Add dependencies to Process Flow components to establish the correct execution order in Process Flows.
• Define Job parameters.
• Add IF - THEN logic to Jobs and Process Flows to ensure the correct conditions exist before they execute.
• Automate retrieval of values from databases to eliminate data entry errors.
• Schedule Jobs and Process Flows to automate production.
Objects
As you build Jobs and Process Flows, you will use a number of objects including:
• Applications to categorize Jobs and Process Flows
• Libraries to specify paths to programs
• Program Types to interface with programs and applications
• Database and Host Logins
• Substitution Variables to store values used in Job parameters
• Queues to control load on your systems
• Output Devices to distribute output
There are many other objects that you may use, or that may be used by the Applications Manager administrator or
operators.
Naming Conventions
You should put some thought into naming conventions for your objects because you cannot readily rename objects
in Applications Manager. This is not a problem if an object is not used in very many places because you can copy
the object, give it a new name, then replace the old object with the new object. But if you have used the object in
many places, copying and replacing is not practical.
Replacing Scripts with Jobs and Process Flows
One of the greatest returns on your investment in Applications Manager can be realized by replacing your scripts
with Applications Manager Jobs and Process Flows. Long scripts that have been used to run nightly batch
processing can be broken up, and each program run by the script can be replaced by a Job. Those Jobs can then
be combined to create a Process Flow that duplicates, or improves, the Job flow in the original script. Elements in
the script that handle output should be replaced by Applications Manager Output Devices.
If the scripts require manual intervention by operators, every attempt should be made to use Applications
Manager predecessors, conditions, and Substitution Variables to automate these manual steps. Predecessors
and conditions can ensure that the Jobs in a Process Flow execute in the correct sequence and that the correct
conditions have been met for the tasks to execute successfully. Substitution Variables can be used to automatically
enter values for parameters where the values are retrieved from your corporate database at the time of execution.
These three features (predecessors, conditions, and Substitution Variables) give you the power to automate
operations to a greater extent than is possible with any other distributed scheduler.
Scheduling
Depending on the size of your organization, you may be responsible for scheduling Jobs and Process Flows, or this
responsibility may fall to production analysts. Either way, Applications Manager has an extensive set of features
for scheduling Jobs and Process Flows. You should be able to create schedules that closely match your corporate
data processing procedures.
Applications Manager 9.4.1
Notice that each step is completed sequentially. The three reports are run as a sub Process Flow, and are
processed simultaneously. However, the GL_IMPORT step will not execute until the three reports have run.
Process Flow Size
How large should you make a Process Flow? There are no hard and fast rules. A Process Flow may contain
up to 999 components. A component can be a Job or another Process Flow. When you add a Process Flow to
another Process Flow, it is considered a sub Process Flow. Sub Process Flows can include their own sub Process
Flows. While you can sub Process Flow to 32 levels, three levels of sub Process Flows is sufficient for most
implementations.
Applications Manager 9.4.1
Below are some guidelines to help you decide how large to make a Process Flow:
• If your entire nightly batch processing centers around one application, and each Job is tightly integrated with the
other Jobs, you could run your entire nightly batch processing from one single Process Flow.
• If you have several processing runs that deal with different applications with no interaction, you might consider
creating a Process Flow for each application. Each Process Flow could be scheduled to run at a set time during
the batch processing window.
• If you are running nightly batch processing that involves multiple applications with many interrelated Jobs, you
might create a number of sub Process Flows, then add those sub Process Flows to one main nightly batch
processing Process Flow scheduled to run at the beginning of your batch processing window.
• If you are modeling a Process Flow that is used in several different batch processing sequences, you may want
to create the flow as a separate Process Flow that you can then add to several other larger Process Flows. This
takes advantage of the fact that Process Flows are reusable objects in Applications Manager. If the Process
Flow changes, you can make edits to the Process Flow in one place and they will take effect everywhere the
Process Flow is in use.
To a great extent, whether you use a few large Process Flows, or create many smaller Process Flows will depend
on the preferences of your operations group. Applications Manager can accommodate both approaches.
Managing Large Process Flows
If you choose to create large Process Flows, sub Process Flows and groups can help you manage them. Both
sub Process Flows and groups can be expanded and collapsed to show and hide their components. Collapsing
sub Process Flows and groups can make it easier to see the big picture in a large Process Flow. The image below
shows a collapsed sub Process Flow called DATA_REPORTS.
In contrast to sub Process Flows that are independent objects added to a Process Flow, groups are created within
a Process Flow from components that already exist in the Process Flow. Groups are used to manage collections
of components. For example, groups can display parallel flows in a Process Flow more clearly. Groups, like sub
Process Flows, can be collapsed to save real estate in the Process Flows window. Unlike sub Process Flows,
groups are not independent objects. They exist only with the Process Flow where they are created and you cannot
copy or reuse them.
You also can add predecessor links to standalone Jobs. In this case, the links identify other tasks that must
complete before the standalone Job will execute.
Internal vs. External Predecessor Links
The Process Flow shown above relies on internal predecessors links, which means all links are contained within
the parent Process Flow, in this case NIGHTLY_PRODUCTION. But what if a step in this Process Flow was
dependent on a step in another Process Flow completing successfully? You would need to create an external
predecessor link as shown below. The link goes from the FTP Job in the NIGHTLY_PRODUCTION Process Flow to
the CK_DISK Job in the External References box.
While creating the Job to run this program, you would include a prompt where the department name could be
entered. The image below shows an FTP Job that takes seven parameters:
Applications Manager 9.4.1
Types of Prompts
There are four types of prompts you can create. The type of prompt determines how the information is entered. The
different types of prompts are described below.
Default The prompt has a default value that users are not
allowed to change. Use this type of prompt when the
parameters for a Job do not change frequently.
Single selection from a list Users may select one (and only one) choice from a
predefined list of possible values. The values can
be pulled from any database table. Use this type of
prompt when end-users will be running the Job from
the Requests window and you want to avoid data entry
errors.
Multiple selection from a list Users may select one or more choices from a
predefined list of possible values. The values can
be pulled from any database table. Use this type of
prompt when end-users will be running the Job from
the Requests window and you want to avoid data entry
errors.
can even define the prompt using a SQL query that will pull a list of values from a database table and present
them to a user when they run the Job from the Requests window. For example, a user could run a sales report
and select their region from a list. By selecting values from a list, you eliminate the possibility of a data entry
error.
• If you add the Job to a Process Flow, prompt values for the Job can be retrieved from the Process Flow header
using special numbered Substitution Variables. This means you can enter one set of values in the Process Flow
header, run the Process Flow using those values, then enter a second set of values and run the Process Flow
again. The entire process can be automated so that the sets of values are pulled from rows in a database table.
The Process Flow is run once for every row in the table. This technique is particularly useful for running Oracle's
Multi-Org.
• The action taken is to delay the task by five minutes and run the check again.
When to Use Conditions
Conditions are an optional feature in Applications Manager. You can create a Process Flow and add Jobs to it, and
the Jobs will execute in the order they are displayed in the Process Flow. Schedule the Process Flow to run on
certain days at a specific time, and you are ready to go. It's quick and easy to build and schedule such a Process
Flow, and it may be all you need.
But you will want to add conditions to the components in a Process Flow if you want to synchronize the Process
Flow with events taking place outside of the Process Flow. For example:
• Run a Process Flow a second time when a certain Job in the Process Flow completes. In other words, the two
runs of the Process Flow will overlap, with the second run starting based on completion of a specific Job in the
first run.
• Run a Process Flow only when a certain number of records exist in a database table.
• Run a Process Flow at 10:00 P.M., but ensure that a specific component in the Process Flow does not run until
after midnight.
• Run a Process Flow only if the Job XYZ completed within the last two days.
Jobs, Process Flows, and Process Flow Components
You can add conditions to standalone Jobs, to Process Flows, and to Jobs and Process Flows within a Process
Flow. This gives you a great deal of flexibility in controlling execution of tasks in Applications Manager.
Before, During, and After Conditions
You can create conditions that are checked before, during, or after a Process Flow or component executes. For
example, for one Job in a Process Flow you might create:
• A BEFORE condition that checks if a specific data file exists.
• A DURING condition that sends an email Notification to operators if the task is running too long.
• An AFTER condition that changes the status from FINISHED to FINISHED WITH ERRORS based on the
return code.
You can add as many conditions as you need to control the execution of Jobs and Process Flows.
Multiple Schedules
You can create more than one schedule for a Job or Process Flow. For example, if you want a Process Flow to run
Monday through Friday at 1:00 A.M., and Saturday at 8:00 A.M., you can create two schedules. You can review the
schedules using the 12 Month Display shown below.
Applications Manager 9.4.1
The INVENT Program Type calls the INVENT interface script, which in turn runs the XYZ task in the Inventory
custom application.
Applications Manager 9.4.1
Applications Manager ships with, or has available as add-ons, Program Types and their matching interface scripts
to run some of the more common programs including:
• Java programs
• SCT Banner
• COBOL
• UNIX and Windows shell scripts
• SQL*Plus programs
• Oracle
• PeopleSoft
If you want to run a custom application, or a third-party software program, you must write an Applications Manager
interface script. The advantage of writing an interface script is that once it is written, you can use it to run an
unlimited number of tasks in the target application or program. This is in keeping with the Applications Manager
object-oriented approach to development.
The Development Guide documents how to create Program Type interface scripts. Applications Manager also
offers consulting services to create the scripts.
Sample UNIX Program Type Script
Below is a listing of a sample EXEC Program Type script used to run UNIX shell scripts. The line numbers have
been added for reference and do not appear in the actual script.
1 :
2 #!/bin/sh
3 #copyright 2009 by Automic Software GmbH
4 # $Header:
/isa/devel/soport/so/dev/sostage/RCS/EXEC-EXECS,v
1.2 2004/02/24 19:06:46 billw Exp $
5 arg="$program `$SQLOPER_HOME/exec/ONELINE $par`"
6 eval $arg
7 err=$?
8 if [ -f $file ]; then
9 $AW_HOME/exec/FILESIZE $file $err
10 err=$?
11 fi
12 if [ -f $OUTPUT/$file ]; then
13 file=$OUTPUT/$file;export file
14 $AW_HOME/exec/FILESIZE $file $err
15 err=$?
16 fi
17 exit $err
The table below lists the key functions a Program Type script must perform, and the corresponding lines in the
EXEC program.
Error determination 7, 10, 15, 17 The code err=$? traps the exit
status of the last command
executed.
Versatility
When you define a Notification, you can enter multiple details covering a variety of conditions. For example, you
can enter one detail that sends out an email when a task finishes. The message could read:
where <Job name>, <date>, and <time> are filled in for each particular task.
Applications Manager 9.4.1
You could enter a second detail that sends out a pager message when a task fails. The message could read:
By using the variables in the messages, the Notification object can apply to any task.
Assigning Notifications
After you have defined a Notification, you can assign it to Process Flow components, Jobs, Applications, and
Program Types. Being able to assign Notification objects to Applications and Program Types makes it easy to set
up Notifications for a whole class of Jobs.
Each Output Scan includes one or more rules. To use an Output Scan, you:
• Define the Output Scan object.
• Add rules to the Output Scan.
• Assign the Output Scan to one or more Jobs and Program Types.
Operations
As an operator, you will monitor tasks as they run through Applications Manager.
Applications Manager provides robust operations tools through the Explorer window. From Explorer, you can:
• Monitor the system.
• Run tasks on an as-needed basis.
• Find out what tasks will run during your shift.
• Troubleshoot tasks.
• Take actions on tasks.
• Handle exceptions to normal processing.
• View and print task output.
• Control load on the system.
• Prevent Applications Manager from launching tasks.
Applications Manager 9.4.1
The Explorer window includes three panes and a status bar. The pane on the left displays a navigational tree.
The item you select in the navigational tree determines what is displayed in the top right pane of the window. The
bottom right pane always displays the History.
The Backlog and History
When a task is submitted in Applications Manager, it is sent to the Backlog. Tasks remain in the Backlog until they
complete successfully or are deleted. Selecting Backlog from the navigational tree displays in the top right pane all
tasks in the Backlog.
Applications Manager 9.4.1
When a task completes executing, Applications Manager removes it from the Backlog and writes a record to the
History. Applications Manager also writes records to the History when tasks are killed or when they fail with a status
such as ABORTED.
Taking the Pulse of the System
The status bar runs across the bottom of the Explorer window and provides status at a glance. Its color reflects
the most severe status of the Applications Manager Automation Engine and Agents, and the tasks running in the
Backlog. The status bar colors have the following meaning:
Color Description
When the Explorer window is minimized, the button on the taskbar uses the same color scheme.
Using the Explorer Tree Icons
The navigation pane on the left side of the window provides a tree structure with selectable object icons. When you
select an icon, Applications Manager displays the matching information in the top right pane of the window. The
pane can show:
• The Backlog (tasks waiting to be processed).
• A filtered list of tasks in the Backlog.
• A summary of objects selected in the object tree.
• Tasks in a particular Process Flow.
Viewing Components in a Process Flow
The Explorer tree displays Process Flow components alphabetically to help you find them. When you select a
Process Flow in the Explorer tree, those same components are displayed in the top right Explorer pane according
to your Backlog search criteria. The default sort method is by task status. To view a flow chart of the Process Flow,
right-click the Process Flow icon and select Predecessors.
Sorting Columns
You can sort any of the columns in the Explorer window to help you find tasks. For example, you can click on the
Status column header in the Backlog to display three different sort orders: ascending (alphanumeric), descending
(alphanumeric), and severity of status. When you sort tasks by severity of status, Applications Manager lists the
most severe task statuses first.
Each scheduled Job/Process Flow includes the start date and time and the Job or Process Flow's name. Process
Flows also include a key icon used to expand/collapse them. To create a basic run book, you can print the current
view of the Forecast window.
Production Schedule
If the forecast does not provide enough detail, you can run a Production Schedule report. Part of a sample
Production Schedule report is shown below.
The report shows all tasks that are scheduled to run, tasks that will be skipped because of the day of the week, and
the conditions assigned to each component in a process flow.
Staging Tasks
A third way to find out which tasks are scheduled to run is to "stage" tasks. By default, the Backlog pane in Explorer
only shows tasks whose run date and time have been reached. Therefore, a task that is scheduled to run an
hour from now will not be displayed in the Backlog. You can override this default behavior by staging tasks to the
Backlog ahead of schedule. To stage tasks, you run the STAGING Job. Whether or not you use the staging feature
will depend on how your operations group functions.
If you are strictly a lights out shop, there is little need to stage tasks. You would not schedule the STAGING Job.
In the rare cases when you need to make changes to a task, you would run the STAGING Job ad hoc for that
task only.
If you are an operations-intensive shop, you may want to stage tasks on regular basis by scheduling the STAGING
Job. You can create a schedule that meets your operations requirements. For example, you could stage all tasks
scheduled to run over the next 12 hours.
runtime information and parameters passed to the task, the file includes error messages. A sample system file is
shown below.
To view system files, you right click the task in the Backlog or History, select Output, and select the system file.
Applications Manager displays the file in the Applications Manager file viewer. From the viewer, you can print the
file.
You can also view task details by right-clicking the task in the Backlog or History and selecting Task Details. Task
details show you various Applications Manager settings associated with a task. From a task's details you can view
the standard out file for a failed task to help determine what went wrong.
Applications Manager 9.4.1
If a task in the Backlog is in a non-running status, you can delete it. For example, tasks with a status of SELF WAIT,
ABORTED or KILLED can be deleted. When you have deleted a task, you cannot reset it directly from the Explorer
window.
Removing All Predecessors for Tasks
If a task in the Backlog is waiting for one or more a predecessors before it can run, you can remove the
predecessor(s) to force it to run.
If a task is scheduled to run at a later time, you can stage it so it is put into the Backlog. Once a task is staged you
can view and edit its task details.
Entering Comments
When you take an action on a task in the Backlog, a comment is automatically created by Applications Manager.
If you wish to include additional information, you can include your own comments to provide relevant information
about the processing of a task and why you made changes to the task.
If you want to run a task at a later time, you can post-date its start date and time.
If end-users submit ad hoc reports, they too can view the output files online, and print or email them. Many
companies use the online viewing feature to replace physical distribution of reports.
How to View Reports
Viewing reports online is as easy as right clicking the task in the History and selecting Output Files. You then can
select the report file or the system file and display it in the File Viewer window as shown above.
Viewing Output in Other Applications
If a task generates output in a format specific to an application such as Microsoft Word or Excel, you can configure
Applications Manager to open the file in that application.
You can define as many Queues as needed to manage the load on your system. For example, you may define:
• A batch Queue that accommodates your nightly batch run.
• A high priority express Queue that handles management report requests.
• A low priority Queue that handles field report requests that are not urgent.
Setting Task and Queue Priority
When you define a task, you can assign it a priority. Applications Manager uses the priority setting to determine
which tasks in a Queue should be executed first.
Likewise, you can assign priorities to Queues. Tasks in a higher priority Queue will be executed before tasks in a
lower priority Queue. To ensure that tasks in a low priority Queue will not be shut out completely, you can reserve
threads on a Queue.
Do You Need to Worry About Queues?
Whether you need to be concerned about Queues and priorities depends on the size of your operation. If you are
a small shop running a couple of dozen tasks a day, one Queue may be all you need. On the other hand, if you are
a large shop running several hundred or several thousand tasks a day, you should spend time determining your
system requirements and defining Queues and priorities. The goal is to maximize use of your computing resources
without overloading them, and to ensure that you meet your service level agreements.
When you idle an Automation Engine or Agent, or inactivate a Queue, currently running tasks will complete, but any
tasks waiting will not start.
Resuming an Automation Engine or Agent
If you want to stop processing newly submitted tasks through an Automation Engine or Agent you can idle it. The
Automation Engine/Agent will go to an Idled status and the icon for the Automation Engine/Agent will be displayed
with a yellow triangle over it in the Explorer tree. Tasks in the Backlog set to run on an Agent in an Idled status will
have a task status of AGENT WAIT. To take the Automation Engine/Agent out of the idle status, you resume it.
Activating Queues
If a Queue is inactive, Applications Manager will still submit scheduled tasks to the Queue but they will have a
status of QUEUE WAIT. When you reactivate the Queue, Applications Manager will process the tasks based on
their priority settings.
5 Installation Guide
The Installation Guide documents the installation procedures for Applications Manager.
The Installation Guide documents the installation procedures for Applications Manager. It is part of the complete
Applications Manager documentation set which includes:
• Getting Started Guide
• Installation Guide
• User Guide
• Development Guide
• Administration Guide
• Oracle Applications Extension Guide
• PeopleSoft Extension Guide
Text Conventions
The following text conventions are used throughout this guide:
• User interface field names, menu items, and window names are written in bold.
• File names and text within scripts are written in bold.
• Variable text is written <within brackets>. In the example below <run ID number> represents the actual run ID
number of a requested Job.
If you submit a large Process Flow, the message will read, 'Task submission in progress: Run ID = <run ID
number>' until all components of the Process Flow have been placed into the Backlog.
Applications Manager 9.4.1
We set jdk.tls.ephemeralDHKeySize to Java version-specific values: 2048 for Java 1.8 and above. If you want
to use 2048 bit certificate keys, both the server (rmiserver) and the client (UI or AgentService) need to be running
Java 1.8 or above. For more information, see your Java documentation.
SSL
The SSL that ships with Applications Manager is the global (exportable) version.
If you want to run with the domestic package, download and unzip the domestic package installation files, then
replace the jar files jcert.jar, jsse.jar, and jnet.jar in the classes directory with the ones from the download. The
classes directory can be found in the following Applications Manager directories:
UNIX:
$AW_HOME/web/classes
Windows:
%AW_HOME%\web\classes
HP-UX Systems
If you are installing Applications Manager on an HP-UX system, the HP-UX Runtime Environment for the Java 2
Platform must be installed on your system prior to running the Applications Manager installation. The installation
must be performed by the HP UNIX administrator with root access. For all other platforms, the Java 2 Platform is
installed by the Applications Manager installation script.
Report Storage Requirements
You will need 300 MB of free disk space on the host before proceeding. There must also be enough disk space to
hold all the output (reports and system listings) from Jobs for the retention period specified. If you will be creating
many large reports, you will need more storage space. The Applications Manager administrator should monitor the
file system to ensure there is adequate space available.
Specifying Ports for Firewalls or Other Purposes
If you have a firewall in place, you will want to specify ports for your Applications Manager processes. During the
installation, you will be asked if you have a firewall in place. If you answer yes, you will be able to specify the port
value(s) at that time. If you are not sure which ports to use, consult your network administrator before installing. For
more information on firewall configuration and port assignments, see Overview of Firewall Settings.
Major Steps
The major steps in the installation process are:
• Create the UNIX account or Windows user.
• Create the Oracle database account.
• Run the installation script.
• Start the supporting servers.
• Start the Automation Engine and Agent processes.
• Verify the installation.
Each step is described in detail in this chapter.
The recommended shells for the account are either KORN or BOURNE shell. The account name can be any
valid UNIX name. In this manual, we represent the UNIX login name as am. You must be logged into the
Applications Manager UNIX account when you perform the install.
2. Pick or create a directory into which you will install Applications Manager.
It is usual practice to set the home directory for the Applications Manager UNIX account (in .profile) to the
install directory for Applications Manager. In this manual, we typically represent the install directory as /home/
users/am.
3. If you will be installing an Automation Engine and Local Agent, ensure that the environment allows Applications
Manager access to the database it will use for its repository by modifying the .profile file for the Applications
Manager UNIX account.
When the Database Is On a Different Machine Than the Automation Engine
If the database is on a different machine than the Applications Manager Automation Engine, you must set the
TWO_TASK variable. To set TWO_TASK, add the following line to the $AW_HOME/site/sosite file where <Oracle
Sid> is the actual name of your Oracle Sid:
Remember, you must log out and log in again after changing any user rights. Otherwise, the changes will not
take effect. You should determine which user(s) the Applications Manager services are going to run as before
proceeding with the installation. You will perform the installation as Administrator.
Special Circumstances
Occasionally, certain Windows host configurations do not allow the Applications Manager WatchWorx service to be
installed using a specific user/password. If you are unable to install the Applications Manager service as a specific
user on your Windows host, Broadcom recommends that you do the following:
Applications Manager 9.4.1
1. Install the Windows Automation Engine/Agent, but leave the Windows User and Password fields blank. By
default Applications Manager installs into the Administrator Account.
2. After you complete the installation, go to the Windows Services and locate the AWW-<Agent name> service.
3. Ensure that the AWW-<Agent name> service's Startup Type option is set to 'Automatic'.
4. Change the Log On As from 'System Account' to 'This Account.' Enter the Windows user account and the
password.
5. Be sure to update the service with an appropriate Windows user name and password.
When the Database Is on a Different Machine than the Automation Engine
If the database is on a different machine than the Applications Manager Automation Engine, you must set the
LOCAL variable. For a new Automation Engine LOCAL should be set in the environment prior to running the
installation. For an existing Automation Engine LOCAL can be set by adding the following line to the %AW_HOME
%\site\sosite.bat file where <Oracle Sid> is the actual name of your Oracle Sid:
LOCAL=<Oracle Sid>
2. To give the Applications Manager account the privileges required to create tables, views, indexes, procedures,
triggers and sequences, issue the following commands:
In the commands above, replace <am>, <tspace>, and <tmpspace> with appropriate values for your system.
3. To prevent the Applications Manager installation script from prompting for the Oracle sys password, you can
grant the appropriate privilege to a system table ahead of time. Make the following grants from the Oracle sys
account:
If you have multiple Automation Engines running in the same database, and you try to install a new Automation
Engine in the same database, you may have trouble granting access to dbms_pipe. If you have trouble, try
stopping the other Automation Engines.
Notes for Oracle 12c and Above
When using Oracle 12c database and you have an Oracle Enterprise Edition multi-tenant system or Oracle
Standard Edition single tenant system, you must install Applications Manager into a pluggable database (PDB)
within a container. Your DBA is responsible for creating the pluggable database.
Oracle 12c and above require the following additional grant:
This is a new requirement as of this release, because the RESOURCE role no longer grants the UNLIMITED
TABLESPACE system privilege by default.
You can change UNLIMITED to 100M or whatever size you feel is appropriate as shown below if space is a
concern.
All the installation related SYS accounts and passwords are for the PDB (each PDB has its own SYS account just
like a database instance prior to 12c).
(Re)granting Access After Rebuilding the Oracle Database
If you rebuild the Oracle database where Applications Manager is installed any time after an initial Applications
Manager installation, you must (re)grant select execute access to the Oracle system table (SYS.V_$SESSION)
and execute to the procedures dbms_pipe, dbms_lock, and dbms_application_info. To do this, you must run the
sysdba.sql script located in the Applications Managersql directory.
To run the script, log into SQL*Plus using the Applications Manager Login and password, then type 'start sysdba'.
The script will prompt you for the sys Oracle password then grant select access to the SYS.V_$SESSION system
table and the procedures. You can also regrant access by executing the SQL statements shown above in Step 3.
Dropping the Database Pipe
If you must drop an Oracle user, your DBA must first drop the database pipe (otherwise you will have to stop and
restart the database before you can reinstall).
Applications Manager 9.4.1
Applications Manager Oracle login none This is the login name, and
name, and password. password to the Applications
Manager Oracle database account
you set up.
Virtual day start time 00:00 This is the time when your daily
processing day starts (usually 00:00
for midnight).
Host name and port for SMTP server none Applications Manager can send out
Notifications of events via email. If
you wish to use this feature, you will
need to enter the host name and
port number of your SMTP server.
Email address for Notifications none The email address you want as the
sender for Notifications.
Automation Engine require its own Yes The Automation Engine must use its
RMI server own RMI server.
RMI registry port 1099 The client uses this port when it
makes initial contact with the RMI
server.
syspassword@pdbname
Parameter Description
RMI Server Port The return port the Agents' AgentService process uses
to communicate to the Automation Engine. This lets
TCP/IP pick random ports each time the processes are
started. This is the typical situation when a firewall is not
in place.
This port is only ever specified in the Automation
Engine's awenv.ini file.
RMI Data Port The listener for the RMI server. Needs to be specified
when a firewall is on the Automation Engine machine.
In UNIX, when you run the installation script, Applications Manager displays the default information for prompts in
brackets [ ]. You can accept the default by pressing the Enter key at the prompt, or type in a response and press
the Enter key.
To run the installation script:
1. Log into the Applications Manager UNIX account.
If necessary, change to the directory where you want Applications Manager installed.
You must be logged into the Applications Manager UNIX account or be logged in as the appropriate Windows
user, and be in the Applications Manager directory for the installation procedure to work correctly. Moving the
files after the installation is complete requires some effort. It is best to install into the correct directory the first
time.
In UNIX, if you transferred files to a staging directory, verify that all the Applications Manager files are owned
by the Applications Manager UNIX login and group. If they are not, the installation will fail. Use the chown and
chgrp commands to modify the ownership if necessary.
2. From the directory in which you want Applications Manager installed, run CDINST.SH on the host.
Be sure to set permissions for the CDINST.SH file.
3. Follow the on-screen prompts using the information from topic System Information Required for the Installation.
4. When presented with the menu of installation options, choose the following option and continue.
. .profile
Applications Manager 9.4.1
The following lines have been added to $HOME/.profile during the installation process:
AW_HOME=/home/<am>;export AW_HOME
. /home/<am>/site/sosite
startso
stopso master
In this series of commands, the startso command starts the awcomm, AgentService, and RmiServer processes
on the Automation Engine.
The stopso master command puts the Automation Engine into a Stopped status, but doesn't stop any processes,
because those elements are performed on threads of the RmiServer process.
Accommodating More than One awcomm Process
If there will be more than one awcomm process running on a host, you must set a different port number for each
process. For information on sharing processes between multiple Agents on the same host, see Installing Multiple
Automation Engines on One Host.
To start the processes from the DOS prompt, issue the startso command. The startso command will start all
applicable Applications Manager processes for the current environment. If an Automation Engine and Local Agent
are installed, startso will start both Automation Engine and Agent processes. If only a Remote Agent is installed,
startso will start only the Agent process.
To stop the processes from the DOS prompt, issue the stopso command. The stopso command will stop all
applicable Applications Manager processes for the current environment. If an Automation Engine and Local Agent
are installed, stopso will stop both Automation Engine and Agent processes. If only a Remote Agent is installed,
stopso will stop only the Agent process.
Starting and Stopping Processes with startso and stopso
The startso and stopso commands are used to start and stop Applications Manager processes. You can issue
them by themselves or with parameters. It is generally recommended that you start and stop processes with the
basic startso and stopso commands, or if monitoring processes with WatchWorx, with the startso watchworx
and stopso watchworx commands. Additional parameters that can be used with the startso and stopso
commands are described in the Administration Guide.
Starting Processes, but not Backlog Tasks
If you are starting Applications Manager as part of a boot sequence, you may want to start Applications Manager
processes, but not allow the Automation Engine to execute tasks until you have had a chance to view the contents
of the Backlog. You can set this up with the following series of commands.
startso
stopso master
In this series of commands, the startso command starts the awcomm, AgentService, and RmiServer processes
on the Automation Engine.
The stopso master command doesn't stop any actual processes, because those elements are performed on
threads of the RmiServer process.
You must download the Applications Manager client from the URL set during the installation your Applications
Manager sent you . Before logging in the first time, you need to configure the client for the Automation Engine(s)
you will be logging into. Once configured, you open an executable .jar file. To log into the client, you must enter
your User name, password, and select an Automation Engine.
Downloading and Opening the Applications Manager Client
To download and open the Applications Manager client:
1. Open your browser and enter the URL set during the installation your Applications Manager sent you. The
screen shown below is displayed.
The default format for the URL is http://<Automation Engine IP address>:<Web Server port number>/
<Automation Engine name>/Intro.html.
For example, if the Automation Engine IP address is 200.2.2.123, the port number is 8080, and the
Automation Engine name is PROD1, the URL would look like the following:
http://200.2.2.123:8080/PROD1/Intro.html
Applications Manager 9.4.1
The Applications Manager client requires a supported version of the Java Runtime Environment (JRE). For a
list of compatible Java versions, click the Compatibility Matrix link.
2. From this page, click the Download the Applications Manager Client link.
This downloads the Client.zip file.
3. Create a directory on your local PC for the client, for example, C:\AM_Client.
4. Move the Client.zip file to the directory you created and unzip the contents in that location.
5. Copy connections.properties.template to connections.properties.
6. Edit connections.properties. Replace {MASTER} with the name of your Applications Manager
Automation Engine and {URL} with the URL to your Applications Manager Automation Engine, such as,
http://127.0.0.1:8080/AMPROD.
For example, you might replace...
{MASTER}={URL}
With...
AMPROD=http://127.0.0.1:8080/AMPROD
To add connection information for additional clients, simply add additional lines in the same format.
7. Copy client.properties.template to client.properties.
8. Optionally, edit client.properties to set client properties.
Text from a sample client.properties file is shown below.
#localOptions=true
#debug=true
#logDir=log
#JAVA_HOME=C:/Program Files/Java/jdk-11
Applications Manager 9.4.1
runOptions=-Xmx256m
#debug=true
It is also possible to set client debug from the About Applications Manager window once you are logged
into the client. However, it is advantageous to set it here, so that debug is on for the client start-up.
• logDir: Allows you to override the name of the sub-directory where client log files are written.
The client log sub-directory includes the following files:
• client.log: Includes all client standard and debug logging up until you click OK on the Logon window.
• <master or connection name>_client.log: Includes all client standard and debug logging for your
client session.
• JAVA_HOME: Allows you to use a different Java than your PC's default Java.
You must use either forward slashes (recommenced) or double backslashes in the path. This is the
standard for Java properties files.
To see your PC's default Java, open a command prompt and type java -version.
• runOptions: Used to increase the memory for your client. Don't make any changes to this setting unless
directed to by Applications Manager support.
• TheuserLanguageFiles: This setting is hidden in the client.properties file and in the documentation by
default. It allows you to specify a master name to use language translations on. When a master is specified,
there will be a drop-down with the languages, which are specified in the Options.properties file for the
master. We believe no one uses this desupported feature anymore.
9. Use RunClient.jar to start the Applications Manager client.
Applications Manager displays the Login window shown below.
Optionally, you may create a shortcut of RunClient.jar and place the shortcut on your desktop.
If the Logon window does not come up, open command window, navigate to the client directory, and enter
java -jar RunClient.jar to see the output. This information will help you or Applications Manager support
troubleshoot the problem.
When starting the local client by double clicking on RunClient.jar (or a shortcut to it) in version 9.3.1 or above,
if it fails to start the error will now be in the startClient.log file.
10. On the Logon window, specify your User, password, automation engine.
User names are not case sensitive, but passwords are.
If you do not know your User name and password, check with your Applications Manager administrator.
11. If the Language field is available, select your language.
Specifying languages for the client is a desupported feature. The Language field is only available in
Applications Manager systems where languages were translated in a previous version.
12. On the Logon window, specify the following:
• The default user name SQLOPER.
Applications Manager 9.4.1
When the Receive RmiServer Error Broadcasts User Option is assigned to a User, any RMI errors will be
displayed in a pop-up window as they occur. Additionally, if any RMI errors occurred since the last time the RMI
server was started, the last ten errors will be displayed in a pop-up window each time the User logs on.
If you have the DBA User Group, you can clear the RMI errors without stopping the RMI server by selecting Clear
RMI Errors from the View menu on the Applications Manager Desktop.
Re-Logging on to Applications Manager
There may be times when you want to log on to Applications Manager under a different user name or connect to
a different Automation Engine. To change your logon from the desktop, go to the File menu and select Re-Login.
You will have to re-enter your password when you re-log in, unless the Disable clearing of login passwords on
re-login Automation Engine option is checked.
Changing Your Password
You can change your password at any time. Your Applications Manager administrator can also set your password
to expire after a certain period of time.
To change your Applications Manager password go to the Options menu on the Applications Manager desktop and
click Change Password.
Accessing the Client Through a Firewall
If you will be accessing the Applications Manager client through a corporate firewall, you must open the appropriate
ports, and specify those ports in the Options.properties file on the host machine.
For details, see Configuring the Applications Manager Client.
For details, see the Installation Guide.
3. Select TEST_JOB from the list and click Request. To quickly find TEST_JOB, type the letter 't' in the Search
field.
The Submit window appears as shown below. TEST_JOB requires one prompt value—a number of seconds to
sleep. It has a default value of 5 (seconds) which will work for now.
4. To run TEST_JOB, click Submit & Close.
Applications Manager 9.4.1
5. To monitor the task, open the Activities menu and select Explorer.
The TEST_JOB should complete successfully with a status of FINISHED. The TEST_JOB will then leave the
Backlog and a record will be displayed in History.
Verification Complete
If TEST_JOB finished, then Applications Manager has been installed correctly. Verification is complete. If you have
special considerations in your environment, you may want to look in chapter Applications Manager Installation—
Advanced Topics for additional information.
If TEST_JOB did not run, or did not complete with a status of FINISHED, contact Broadcom Support.
Applications Manager 9.4.1
$AW_HOME/web/classes
Windows:
%AW_HOME%\web\classes
Options.properties
The Options.properties configuration file controls how the Applications Manager clients interact with the
Automation Engine, and other features such as the image used for the splash screen. The file includes host,
database, and login information. The file is self-documenting. However, if you have questions about altering the file,
contact Braodcom Support.
Accessing the Client through a Firewall
If you will be accessing the Applications Manager client through a corporate firewall, you must open the appropriate
ports, and modify the Options.properties file on the host machine. For details, see Overview of Firewall Settings.
When Applications Manager creates the OS user, it does not assign it a password, because you should not have to
log into Applications Manager as the OS user.
If you must log into Applications Manager as the OS user, go to the Users window, select the OS user, and create
a password as shown below.
Applications Manager 9.4.1
bin/CLASSPATH
/web/classes/AppWorxLang.jar
Applications Manager 9.4.1
/web/classes/UserWorx.jar
ORCLFAILOVER =
(DESCRIPTION_LIST =
(FAILOVER = TRUE)
(LOAD_BALANCE = FALSE)
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)
(HOST = vip-linux1)
(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = awrac.am.com)
(INSTANCE_NAME = awrac1) ) )
(DESCRIPTION =
(ADDRESS = (PROTOCOL = TCP)
(HOST = vip-linux2)
(PORT = 1521))
(CONNECT_DATA =
(SERVER = DEDICATED)
(SERVICE_NAME = awrac.am.com)
(INSTANCE_NAME = awrac2) ) ) )
$AW_HOME/web/classes
Applications Manager 9.4.1
Windows:
%AW_HOME%\web\classes
The entry is "OracleRAC" and it is nearly identical in content to the tnsnames.ora entry. It should contain all the
information for the database instances that Applications Manager can connect to. Here is an example of the same
connection above as it would appear in Options.properties:
OracleRAC= (DESCRIPTION = \
(ADDRESS = (PROTOCOL = TCP) \
(HOST = vip-linux1) \
(PORT = 1521)) \
(ADDRESS = (PROTOCOL = TCP) \
(HOST = vip-linux2) \
(PORT = 1521)) \
(LOAD_BALANCE = FALSE)\
(FAILOVER = TRUE) \
(CONNECT_DATA = (SERVER = DEDICATED)\
(SERVICE_NAME = awrac.am.com) ) )
Note that the slashes at the end of each line are required so the entry is interpreted by the RMI Server as one
continuous line.
Also, setting the OracleRAC line will override the DB_IP, DB_PORT, and DB_SID in the awenv.ini file.
Notes on Behavior
At this time we recommend using Applications Manager only in a FAILOVER configuration for Oracle RAC. Load
balancing configuration should be avoided.
Our processes 'sleep' and 'wake up' on Oracle DMBS pipes. Pipes are instance-specific, so if Applications
Manager processes are balanced across instances, then wake-ups issued by one process may not reach the
target process. The Automation Engine and Agent sleep times can be reduced, but this is not ideal because of
performance issues.
Expected Failover Behavior
When a failover happens (one instance goes down), the Applications Manager processes should connect to the
new Oracle instance fairly quickly. Because some of the database processes might have been issuing a database
transaction at the time of the failure, it is possible to lose a transaction or two during this switch. This could result in
a task going into a DIED status or a node going into BUSY status.
We have not seen this in our testing, but it is a possibility. Our processes will retry most statements if they get a
failure on the first try.
If these types of errors do happen, it is expected behavior and not critical. For died tasks, check the task output to
see that the process was finished (it should have been). Agents that go into a BUSY status should go back to a
RUNNING status at the end of the sleep cycle after the Applications Manager stack has reconnected.
• Web server
• RMI registry
• RMI data port
When installing multiple Automation Engines on one host, you will need to pay careful attention to your
environment. For example, on UNIX you need to consider how you call sosite. When sosite is called in the UNIX
user's .profile script, that means the wrong sosite might be invoked. Different situations call for different solutions,
but consider running Automation Engines under different OS users or invoking the sosite file manually, not in
.profile.
The awcomm Process
The awcomm process provides a port directory service for the Automation Engine machine. You need only one
awcomm process, even if you are running different Applications Manager instances and versions. This is possible
because you can shut down an Applications Manager instance without shutting down the awcomm process.
The default setting for the awcomm port is 2136. To change the awcomm port, edit the AWCOMM_PORT line in
the [default] section of the awenv.ini file in the site directory.
RMI Registry Port
The default setting for the RMIRegistryPortNumber is 1099. The file you need to edit to change the port is listed
below.
The Options.properties file in the following directory:
UNIX:
$AW_HOME/web/classes
Windows:
%AW_HOME%\web\classes
RMIDataPortNumber Port
The default setting for the RMIDataPortNumber is 0. The actual port used will be a randomly chosen open port.
You can keep this at 0 for all your Automation Engines, if you want (as long as you don't have a firewall). The file
you need to edit to change the port is listed below.
The Options.properties file in the following directory:
UNIX:
$AW_HOME/web/classes
Windows:
%AW_HOME%\web\classes
RMI server process: The RMI server process is a Java process that runs on the Automation Engine machine. The
RMI server IP address is the only IP that is specifically set during the install. This setting, RMIHostID, is specified in
the Options.properties file.
The location for the Options.properties file is:
UNIX:
$AW_HOME/web/classes
Windows:
%AW_HOME%\web\classes
The Automation Engine process runs as a thread under the Java process that also runs the RMI server. It will use
the same IP address that the RMI server uses, which is the RMIHostID setting.
Agent service process: The Agent service process is a Java process that runs on every Agent. The Agent service
by default uses the IP address returned by the gethostbyname function. You can see what this is by executing
the gethost program in the c directory on the Agent. This will be the return IP address that the Automation Engine
machine will use to communicate back to on the Agent. It will also be used for validation and should match the IP
address that is set in the client's Agent manager when creating the entry for the Remote Agent.
If gethost does not return the correct IP, it needs to be set manually. To set this manually to a different IP address
you can set the variable AgentIP in the awenv.ini file.
awapi process: Each Agent has an awapi process that connects to the Local Agent service process. In order for it
to connect successfully, the awapi process must be validated by the Agent server process. To do this, it compares
the Applications Manager OS user and IP address that the awapi process starts under to the OS user and IP
address that were set when creating the Agent object.
The awapi process will also use the IP address returned by the gethostbyname function, which you can determine
by executing the gethost program in the c directory on the Agent. This cannot be changed. However, if this does
not match what is in the Agent definition, you can create a file named remote_rmi.dat in the data directory on the
Agent to allow the Agent service to validate the awapi. This file should contain one line with the values:
So, for example, it would look like this if the IP address in the Agent definition is set to '200.1.1.1':
Sample Configuration
Here is an example of the settings for a network with multiple IP addresses:
Automation engine machine:
• Automation engine name: MSTR_TEST
• IP addresses: 200.1.1.1, 200.1.1.2
• IP address for clients to access: 200.1.1.1
• IP address used for communicating to Agents and that Agents will communicate to: 200.1.1.2
• IP Address returned by gethost: 200.1.1.2
Remote Agent machine:
• Agent name: RA_TEST
Applications Manager 9.4.1
• User: awuser
• IP addresses: 200.1.1.3, 200.1.1.4
• IP address set in the Agent definition: 200.1.1.3
• IP address returned by gethost: 127.0.0.1
Automation engine settings:
The Options.properties file will have the following setting:
RMIHostID=200.1.1.1
[default]
RMIServerIP=200.1.1.2
[default]
AgentIP=200.1.1.3
UNIX:
$AW_HOME/web/classes
Windows:
%AW_HOME%\web\classes
• RmiRegistryPortNumber <-> Applications Manager client: Activity will occur when the client first logs in,
then this connection is closed. Activity will occur on the entry of data and during the refresh process. This
connection is dependent on auto-refresh being on if no interaction is taking place.
• The Applications Manager Automation Engine runs on an IP address, but is recognized as having another IP
address.
• For failover, you may need to list the IP addresses of multiple Automation Engines, Agents, and/or clusters that
are used by a Remote Agent.
When the Automation Engine communicates with the Remote Agent, the incoming IP address is validated. To
validate the Automation Engine's IP address, the Remote Agent checks its awenv.ini file in the site directory for
the IP address listed on the MASTER_IP_ADDRESS line. If that IP address does not validate, the Agent checks
the IP address listed on the ALT_MASTER_IP_ADDRESS line (if present). Sample code from the awenv.ini is
shown below:
MASTER_IP_ADDRESS=400.3.2.45
ALT_MASTER_IP_ADDRESS=444.3.2.66
keytool -keystore user_keystore -keyalg RSA -genkey -alias "AM" -storetype JKS -storepass
<password>
The following is a sample location where the file gets generated: C:\Program Files\AdoptOpenJDK
\jdk-11.0.6.10-hotspot\bin
• With CA issued certificate:
A .CER file can be imported to a keystore using the following command:
keytool -importcert -file certificate.cer -keystore user_keystore -alias "AM" -storetype JKS -
storepass <password> -trustcacerts
The following is a sample location where the file gets generated: AW_HOME\data
CA Issued Certificate
From 9.3.5 and above, if the Certificate is CA Issued Certificate, copy the generated user_keystore and
user_keystore_config files to the <install-dir>\data directory present on the Automation Engine
machine.
Applications Manager 9.4.1
If the certificate is self-signed, user_keystore and user_keystore_config files need to copied to Remote
Agents and Client machines.
On each user's client machine, create a C:\Users\<user name>\AppWorx\<master name> folder for each
master in the connections.properties file where <user name> is the actual user's name and <master
name> is the name of the master. Then place copies of the user_keystore and user_keystore_config
files for each master in the sub-directory for that master. This allows for different keystores to be used on each
master.On each Remote Agent machine, the user_keystore and user_keystore_config files need to be
copied to data directory of the Remote Agent installation directory.
. sosite
Windows:
sosite
stopso all
stopso awcomm
rm -r $AW_HOME
The WatchWorx service monitors Applications Manager processes and restarts them if they stop. You should
remove the service before you remove the Applications Manager directory. To remove the WatchWorx service on a
Windows system, run the Applications Manager installation script awinstall and choose option 6.
Removing the AW_HOME Directory in Windows
When you installed Applications Manager, you designated a home directory. You must remove this directory and all
sub directories and files.
To remove the directory:
1. Using Windows Explorer, navigate to the parent directory of the Applications Manager home directory.
2. Delete the directory.
Dropping the Oracle User
When you installed Applications Manager, you created an Applications Manager Oracle user, usually 'am' or
something similar. You must drop the user and all tables associated with the user.
If it becomes necessary to drop an Applications Manager Oracle user, you must first run the following SQL*Plus
script logged in as that particular Oracle user (otherwise you will have to stop and restart the database before you
can reinstall):
Where <user name> is the name of the Oracle user and is not SYS or SYSTEM.
To drop the user, log into Oracle as SYS and issue the following command:
You must have AW_RMI_ONLY set in the sosite file. This is set automatically by the installation script in order
to prevent the start of an AgentService process.
If you ever want to add a Local Agent to the RMI server, you can re-run the installation script, and enter Y for the
Install Automation Engine's Agent prompt.
2. In the awenv.ini file for each AgentService, add the line for the failover RMI server:
alt_master_ip_address=<IP address>
3. Specify the new RMI server as an allowable RMI server by designating it as a master on the RMI Servers tab of
the Connections window in the Applications Manager client.
The Connections window is limited to Users who have the DBA User Group. If you do not have access to it,
see your Applications Manager administrator.
In the image above, the Automation Engine is shown at the top of the table. Its session ID is 3025715. Its
corresponding RMI server (RmiServer) has the same session ID. The Automation Engine's Local Agent is
named V9AM01 and displays the same IP address as the Automation Engine's RMI server. The final item is the
second RMI server.
The last column in the table, Master, applies only to the RMI servers. If "Ok" is displayed in the column, it
indicates that the RMI server can support the Automation Engine.
Applications Manager 9.4.1
To activate an RMI server to serve the Automation Engine, double-click the entry in the table. This will toggle the
entry from "No" to "Ok". To toggle the entry back to "No", double-click it again.
RMI Failover Behavior
The Applications Manager Automation Engine supports active-passive failover. A secondary Applications
Manager RMI server can be setup and run in stand-by mode (active-passive), so that in the event that the primary
Applications Manager Automation Engine goes down, the stand-by RMI server will become the active Automation
Engine. When the stand-by RMI server becomes active, the Agents will reconnect to it automatically; with the
exception of the Local Agent. That is because the Local Agent is part of the Automation Engine installation, and
integrated with the specific Automation Engine system. Therefore, in systems that require RMI server failover, it is
recommended that all Jobs be setup to run on Remote Agents.
The name can be up to 30 characters long. If you install two or more Remote Agents on a machine, do not
give them the same name (even if they report to different Automation Engine instances). The following are
reserved words that cannot be used for Automation Engine or Agent names: ALL, APPWORX, AWCOMM,
AGENTSERVICE, AWAPI, RMI, RMISERVER, AGENT, and MASTER.
• Description
A description of an Agent can be up to 30 characters long.
• IP address
The IP address or DNS name.
• CPU Limit
Defines the percentage of CPU usage where Applications Manager will spawn no new tasks. When an Agent
reaches its CPU limit it will go into a CPU WAIT Agent status. Tasks waiting to run on an Agent that has
reached its CPU limit will go to an AGENT WAIT status. The default is 80% for newly created Agents or 99%
for Agents that were upgraded from a version of Applications Manager before this feature existed. You can
see actual CPU usage percentages for Agents in the Cpu column on the Agent Summary on the Explorer
window. The CPU usage is updated about once every minute for each Agent and the Automation Engine.
• Thread Schedule
Sets the maximum number of concurrent tasks that can run on the Agent at one time in all Queues.
Editing the Thread Schedule of the Automation Engine/Local Agent from this field will change the setting for
only the Local Agent.
Thread Schedules can be changed for the Automation Engine and Agents from the Explorer window.
• OS type
Indicates that the operating system is UNIX.
• User
The operating system user for the Agent. The User selected determines the Awexe range assigned to the
Agent. Be sure your Agents are assigned to a User with the full Awexe range, which is 1000-9999. Users
assigned to a Remote Agent require an Awexe range for the Agent to start.
Warning: This User should never be SQLOPER.
1. Log into the Applications Manager UNIX account on the remote host, and if necessary, change to the directory
where you want the Applications Manager Remote Agent installed.
You must be logged into the Applications Manager UNIX account and be in the Applications Manager directory
for the installation procedure to work correctly. Moving the files after the installation is complete requires some
effort. It is best to install into the correct directory in the first place.
If you transferred files to a staging directory, verify that all the Applications Manager files are owned by the
Applications Manager UNIX login and group. If they are not, the installation will fail. Use the chown and chgrp
commands to modify the ownership if necessary.
2. From the directory in which you want the Agent installed, run CDINST.SH on the host.
Be sure to set permissions for the CDINST.SH file.
After completing the initial install, you can run agentinstall and bypass cdinst by typing:
sh agentinstall
The agentinstall program can be used to modify the current agent-related parameters for the Applications
Manager installation.
3. Follow the onscreen prompts.
4. After completing the installation script, enter the command as shown below to establish the proper environment,
or log out of UNIX and back in.
. .profile
The following lines have been added to $AW_HOME/.profile during the installation process:
AW_HOME=/apps/appworx_a;export AW_HOME
. /apps/appworx_a/site/sosite
* There must be enough disk space to hold all the output (reports and system listings) for the retention period of the
Job.
Accessing the Remote Agent Through a Firewall
When running the Remote Agent install script, you will be asked if you have a firewall installed. If you answer yes,
you will be able to specify values for the following ports:
• Agent Server Port
• Agent Client Port
• Stack Server Port
• Stack Client Port
It is a good idea to specify the same port numbers on each Remote Agent to minimize the number of ports you
have to keep open on the Automation Engine's firewall.
After specifying port numbers for the Remote Agent (as well as for the Automation Engine), you must open the
appropriate ports in your firewall software on the Automation Engine/RMI server, Agent, and client. If you need to
change your firewall ports, you can do so by re-running the install script. For more information on the ports you
need to open for inbound and outbound firewalls, see Overview of Firewall Settings.
Administrator (recommended) or Applications Manager • Advanced User Right: act as part of the operating
system
• Advanced User Right: log on as a service
• File access permissions to run the programs you
wish to run with this Automation Engine and Local
Agent.
Applications Manager 9.4.1
Remember, you must log out and log in again after changing any user rights. Otherwise, the changes will not
take effect. You should determine which user(s) the Applications Manager services are going to run as before
proceeding with the installation. You will perform the installation as Administrator.
Special Circumstances
Occasionally, certain Windows host configurations do not allow the Applications Manager services to be installed to
a specific User/Password. If you are unable to install the Applications Manager services as a specific user on your
Windows host, Applications Manager recommends that you do the following:
1. Install Windows Remote Agent and Network Listener Services; however, leave the User and Password fields
blank. By default Applications Manager installs into the System user account.
2. After you complete the installation, go to the Windows Services and locate the AWW-<Agent name> service.
3. Ensure that the AWW-<Agent name> service's Startup Type option is set to 'Automatic'.
4. Change the Log On As from 'System Account' to 'This Account.' Enter the Windows user account and the
password.
5. Be sure to update the service with an appropriate Windows user name and password.
1. Backup {Tomcat_home}/conf/server.xml.
Applications Manager 9.4.1
3. Add <Context path> attribute where path={AM Master} and docBase={location of AM web
folder}
<Context path="/AM_DEV"
docBase="C:/AM_DEV/web/"
reloadable="true"
crossContext="true"/>
Before you begin the upgrade, stop the Applications Manager processes and back up the Applications Manager
database account.
Make sure you have the following information before proceeding:
• The Applications Manager UNIX account login and password or the Applications Manager Windows
administrator login.
• The Applications Manager Database Login and password
Procedure
To upgrade an Applications Manager Automation Engine and Local Agent:
1. Stop all Applications Manager processes, including awcomm. You can kill all Applications Manager processes
by issuing the following commands:
stopso all
stopso awcomm
If you have multiple RMI servers installed for RMI failover, you must stop the RmiServer processes on all
machines where they are running.
2. Log into the Applications Manager account, and if necessary, change to the directory where Applications
Manager is installed.
In UNIX, you must be logged into the Applications Manager UNIX account and be in the Applications Manager
directory for the installation procedure to work correctly. If you transferred files to a staging directory, verify
that all the Applications Manager files are owned by the Applications Manager UNIX login and group before
running the install script. If they are not, the upgrade will fail. Use the chown and chgrp commands to modify
the ownership if necessary.
3. From the directory in which you want Applications Manager installed, run the installation script. In UNIX, the
script is CDINST.SH. In Windows, it is cdinstall.bat.
4. Follow the onscreen prompts.
Note that the required information for an install/upgrade may have changed since you first installed Applications
Manager. For more information on the current requirements, see System Information Required for the
Installation.
5. When presented with the menu of upgrade options, choose:
. .profile
chmod 4711 c/SURUN
chown root c/SURUN
Applications Manager 9.4.1
2. If you are using surun, make sure you move the surun file to the SURUN directory. From the User account
$AW_HOME/c directory, type:
mv c/surun c/SURUN
CDINST.SH
Applications Manager 9.4.1
Windows:
cdinstall.bat
. .profile
chmod 4711 c/SURUN
chown root c/SURUN
2. If you are using surun, make sure you move the surun file to the SURUN directory. From the User account
$AW_HOME/c directory, type:
mv c/surun c/SURUN
If TEST_JOB finished and you can view the output file, then Applications Manager has been upgraded correctly.
Verification is complete.
2. From the Rapid Automation Loader window, browse to a directory where RA components are on your PC
using the Browse button.
3. Select a component from the Available Components box.
4. Optionally enter an ID number for the component in the ID field. If you don’t enter an ID number the next
available number will be used.
In most cases you would not need to worry about ID numbers. You would only need to set one in rare cases
such as when an Agent with references is deleted out of the database and it needs to be recreated with the
same number in the SO_OPERATORS table.
5. Click OK or Apply.
OK saves the changes and closes the window. Apply saves the changes and keeps the window open. In either
case a confirmation window will tell you that the component was loaded and the component will be added to the
Loaded Components box.
You will now be able to define Agents of this type in Applications Manager.
6. If you are upgrading an existing RA Agent type that you have Agents defined for, you must stop and restart the
AgentService process for all previously defined Agents.
7. For some Agents, such as Business Objects, you must log out and log back in to the Applications Manager
Client.
Deleting Loaded Components
To delete a loaded component, select the component and click Delete. In order to delete a component, you must
first delete any objects from Applications Manager that reference this component.
Comparing Build Information
To compare build information between available and loaded components, select an available component from
your PC in the Available Components box and a loaded component from the Loaded Components box and
click Build Info. Applications Manager opens the Build Info window shown below where you can compare the
information for the two builds.
Applications Manager 9.4.1
$AW_HOME/web/classes_alt
Windows:
%AW_HOME%\web\classes_alt
After adding additional files to the classes_alt directory, you must stop and restart process by going to the site
directory and issuing the following commands:
UNIX:
. sosite
stopso
startso
Windows:
sosite
stopso
startso
Applications Manager 9.4.1
6 User Guide
The User Guide is a comprehensive procedures manual that covers all aspects of Applications
Manager operations.
The User Guide is a comprehensive procedures manual that covers all aspects of Applications Manager
operations. It is part of the complete Applications Manager documentation set which includes:
• Getting Started Guide
• Installation Guide
• User Guide
• Development Guide
• Administration Guide
• Oracle Applications Extension Guide
• PeopleSoft Extension Guide
Text Conventions
The following text conventions are used throughout this guide:
• User interface field names, menu items, and window names are written in bold.
• File names and text within scripts are written in bold.
• Variable text is written <within brackets>. In the example below <run ID number> represents the actual run ID
number of a requested Job.
If you submit a large Process Flow, the message will read, 'Task submission in progress: Run ID = <run ID
number>' until all components of the Process Flow have been placed into the Backlog.
Viewing Forecasts
Using the forecast feature, you can view a list of scheduled Jobs and Process Flows.
Requesting Jobs and Process Flows
There may be times when you want to run tasks outside of a set schedule. You can submit individual Jobs and
Process Flows from the Requests window. In Applications Manager, we refer to these as "ad hoc" requests.
Using Gantt Windows
You can use the following operations Gantt windows:
Applications Manager 9.4.1
• Graphical Forecast: Displays scheduled Jobs and Process Flows in a Gantt chart format.
• Backlog Gantt View: Displays the contents of the Backlog in a real-time Gantt chart format. You can take
actions on tasks or view/edit their task details.
• History Gantt View: Displays in Gantt chart format how the components of a Process Flow in the History
executed.
Viewing Operations Reports
Applications Manager comes with a set of predefined Reports that provide information about your Applications
Manager objects. You can also import an extensive set of Applications Manager History Analysis Reports that you
can use to review how tasks were processed. You can also create your own custom Reports.
If you are using a Windows client with 800x600 resolution, you will need to select the Windows auto hide taskbar
option.
ToolTips
ToolTips provide a brief description of buttons, icons, and fields. To see a ToolTip, rest the mouse pointer over the
button, icon, or field. A ToolTip appears after the mouse pointer has remained motionless for a second or two. In
the image above, the mouse pointer is resting on the Jobs icon in the toolbar. You can disable ToolTips by going to
the View menu and unchecking the ToolTips option.
Toolbar and Menus
Applications Manager 9.4.1
The toolbar consists of a row of icons running across the top of the screen. Click an icon to open its corresponding
window. You can also access the windows in the toolbar from the Operations and Object Admin menu items. The
Activities menu listings open windows where you can take actions such as opening Explorer or running ad hoc
tasks with Requests. The Object Admin menu listings open selector windows where you can view, add, edit, or
delete object definitions (depending on your User Group access).
You can view or hide the toolbar by opening the View menu and checking/unchecking the Toolbar option. You can
optionally view text for the toolbar options by opening the View menu and checking/unchecking the Toolbar Text
option. You can add or remove the icons displayed on the toolbar by selecting Settings from the Options menu.
For instructions on editing desktop settings, see Editing General Desktop and ToolBar Settings.
Taskbar
The taskbar is a graphic bar running across the bottom of the desktop that is used to select active windows. When
you open an Applications Manager editing window, the window is represented by an icon in the taskbar. From the
taskbar, you can right-click a window icon to:
• Restore a window to the desktop or minimize it to the taskbar. You can also display an active window by
opening the View menu, selecting Windows, and choosing a window.
• Maximize a window to fill the desktop.
• Move a window to the front of the desktop.
• Close a window.
Selector windows are used when defining Applications Manager objects. They are not displayed on the taskbar
because they do not contain unique information and are represented by icons on the toolbar.
Status Bar
The status bar is displayed across the bottom of the Explorer window. Its color alerts you to the status of the
Automation Engine, Agents, and tasks running in the Backlog. When the Explorer window is minimized it uses the
same color scheme on the taskbar. For more information on the status bar, see Monitoring with the Status Bar and
Object Icons.
Closing All Windows or Selector Windows
To close all windows, go to the View menu and select Close all. To close only the selector windows, select Close
selectors.
These settings are not retained when you close the window. Many columns can be configured and saved. For more
information, see Customizing Tables.
Copying, Cutting, and Pasting Text
You can right-click in a field to bring up a pop-up menu with copy, cut, and paste options.
Bringing Error Dialogs into View
If you minimize the Applications Manager desktop when an error message is displayed, the Applications Manager
desktop may seem to lock up. This is because the error dialog is out of view. To bring the error dialog back into
view, hold down the Alt key and press the Tab key until you highlight the Java coffee cup icon.
Keyboard Navigation
You can use the following keyboard navigation in Applications Manager:
• Ctrl-Up arrow key sets the focus on the current tab, then the right and left arrow keys will navigate to the other
tabs. Ctrl-Down arrow key returns focus to the page.
• If a table has focus, then Ctrl-Tab will move the focus out of the table.
• When you are in a table, Enter and Tab are used for navigation within the table.
• To alternately expand and collapse an object's key, press Enter.
• To select a mnemonic key, click Alt+<the key>.
Assigning Options
When required, you assign options and objects using the type of window shown below.
Applications Manager 9.4.1
Assign objects by moving them from the Unassigned column to the Assigned column. The table below describes
how to assign multiple options.
To: Do this:
Move a selected value between the two columns Double-click the value.
-or-
Select the value and click the single arrow button.
Move all values between the two columns Click on the double arrow button.
Make multiple contiguous selections Hold down the Shift key and click the first and last
values.
Make multiple nonadjacent selections Hold down the Control key and click on each value.
4. You can set alerts in Applications Manager that are triggered by the status of tasks in the Backlog by selecting
the Alerts tab. For example, you can play a sound whenever a task aborts. For more information, see Setting
Alerts.
5. Applications Manager can log status changes for tasks, Agents, and the Automation Engine. If you have the
DBA User Group, you can configure status logging options from the Task Events and Agent Events tabs of the
Settings window. If you do not have the DBA User Group, you will not see these tabs.
6. To save the settings for the current session and future sessions, click OK.
What Are User Options?
User Options are additional settings selected by your Applications Manager administrator. They control User
access to Applications Manager features such as the Hide feature in the Output window.
Setting Alerts
If the color change in the status bar is not a sufficient alert for your monitoring purposes, you can set additional
alerts for specific statuses. For example, you might have a sound or voice message play when a task aborts. Alerts
launch a program or script on the client machine while you are logged on. To set the alerts, use the Alerts tab on
the Settings window shown below.
Statuses
You can define alerts for the following statuses:
• Aborted task: when a task completes with a non-FINISHED status such as ABORTED
• Hold task: when a task goes into a HOLD status
• Agent trouble: when an Agent goes into a status that requires operator intervention such as a BUSY or
TROUBLE
• All clear: when all task/Agent statuses are changed so that tasks are eligible to run again
You can choose to trigger the alert when the first task or Agent displays the status, or every time a new task or
Agent displays the status.
Local Windows Programs
Applications Manager 9.4.1
An alert runs a local Windows program that you specify in the field next to the alert category. You must enter a
command which includes the full path to the program file. In the image above, commands like the following are
used to play .wav files with the operating system's default player:
cmd /c="C:\Sounds\abort_task.wav"
You can view Reports for each of the operations windows and selector windows.
An operations window Open the operations window and choose the Report
type from the Reports menu.
This opens the Reports window and selects the Report type corresponding to the window you opened it from. In
the image below, the Reports button is selected on the Jobs Selector window, opening the Reports window with
the Jobs type highlighted.
To view Reports for another object type, select that object type from the Type box. If an object is not listed in the
Type box, there are no Reports for it. Once you select a Report, click the Show button.
Applications Manager 9.4.1
You can specify the number of lines displayed on each page using the Lines per Page field. The new setting will
go into effect when you click the Redisplay button. Doing so will update the time and date in the Report header, but
not the data displayed in the Report.
Running Applications Manager History Analysis Reports
You can import Applications Manager History Analysis or AHA Reports. They include an AHA prefix designation in
their names. For example, the AHA-FINISHED_JOBS_BY_STATUS_HR Report returns completed tasks according
to status by hour. Data relating to AHA Reports is generated and loaded into Applications Manager by running the
CALC_HISTORY_STATISTICS Job.
Customizing Tables
Many of the Applications Manager windows display tables of information. In these windows, you can choose the
columns displayed and their order using Setup windows.
Process Flow Summary The Process Flow Summary on the Explorer window.
Task Output files The Output files tab on the Task Details window.
Gantt task summary The information displayed in the pop-up table when you
hover over a Job or Process Flow in the Backlog Gantt
view window and the Flow Diagram window.
Backlog task summary The information displayed in the pop-up table when you
hover over the Run ID column for tasks in the Backlog.
The steps for choosing columns and changing the column order are the same for all tables. To edit a table,
open the Options menu, select Tables, then select the table you want to edit. The Setup window for the Queue
Summary table is shown above.
The top of the window shows the table as it will be displayed in Applications Manager. The bottom of the window
displays a list of the columns that can be displayed.
Making Changes
The table below describes how to customize Applications Manager tables.
To: Do this:
Change the name of a column Edit the entry in the Name column.
Control the width of a column Enter a number of pixels in the MAX_WIDTH and
MIN_WIDTH columns. If you enter MIN_WIDTH
settings, its possible that some columns will be pushed
off the table. If this happens, select the Horizontal
scrolling option described below.
Format date columns Select a date format from the FORMAT column.
Control the margins within a column In the Column marginfield, enter a number of pixels to
be added to the left and right margins of the column.
Change the order of the columns Select the title of a column in the top window and drag it
to a new location.
Applications Manager 9.4.1
To: Do this:
Return the table to its last saved setting Click the Reset button at the bottom of the window.
Default Settings
Your Applications Manager administrator can set default settings for Applications Manager tables. If you customize
tables, the default settings are overridden.
Procedure
To change the color for a group of statuses:
1. From the Options menu, select Status colors.
Applications Manager 9.4.1
Group Statuses
To view copyrights for third party software used by Applications Manager, click Copyrights. Copyright information
also exists in the copyrights directory on the Applications Manager Automation Engine.
Java Garbage Collecting from the Applications Manager Client
You can do a Java garbage collection to free up heap memory that is no longer needed by clicking Java Garbage
Collection under the Debug menu on the About Applications Manager window. This will only send a hint to
Java, so it doesn’t force it to collect everything.
Setting Debug from the Applications Manager Client
You can turn on debug for the Applications Manager client, the RMI server, or Oracle trace by selecting options in
the Debug menu.
http://200.2.2.123:8080/PROD1/Intro.html
The Applications Manager client requires a supported version of the Java Runtime Environment (JRE). For a
list of compatible Java versions, click the Compatibility Matrix link.
2. From this page, click the Download the Applications Manager Client link.
This downloads the Client.zip file.
3. Create a directory on your local PC for the client, for example, C:\AM_Client.
Applications Manager 9.4.1
4. Move the Client.zip file to the directory you created and unzip the contents in that location.
5. Copy connections.properties.template to connections.properties.
6. Edit connections.properties. Replace {MASTER} with the name of your Applications Manager
Automation Engine and {URL} with the URL to your Applications Manager Automation Engine, such as,
http://127.0.0.1:8080/AMPROD.
For example, you might replace...
{MASTER}={URL}
With...
AMPROD=http://127.0.0.1:8080/AMPROD
To add connection information for additional clients, simply add additional lines in the same format.
7. Copy client.properties.template to client.properties.
8. Optionally, edit client.properties to set client properties.
Text from a sample client.properties file is shown below.
#localOptions=true
#debug=true
#logDir=log
#JAVA_HOME=C:/Program Files/Java/jdk-11
runOptions=-Xmx256m
#debug=true
It is also possible to set client debug from the About Applications Manager window once you are logged
into the client. However, it is advantageous to set it here, so that debug is on for the client start-up.
• logDir: Allows you to override the name of the sub-directory where client log files are written.
The client log sub-directory includes the following files:
• client.log: Includes all client standard and debug logging up until you click OK on the Logon window.
• <master or connection name>_client.log: Includes all client standard and debug logging for your
client session.
• JAVA_HOME: Allows you to use a different Java than your PC's default Java.
Applications Manager 9.4.1
You must use either forward slashes (recommenced) or double backslashes in the path. This is the
standard for Java properties files.
To see your PC's default Java, open a command prompt and type java -version.
• runOptions: Used to increase the memory for your client. Don't make any changes to this setting unless
directed to by Applications Manager support.
• TheuserLanguageFiles: This setting is hidden in the client.properties file and in the documentation by
default. It allows you to specify a master name to use language translations on. When a master is specified,
there will be a drop-down with the languages, which are specified in the Options.properties file for the
master. We believe no one uses this desupported feature anymore.
9. Use RunClient.jar to start the Applications Manager client.
Applications Manager displays the Login window shown below.
Optionally, you may create a shortcut of RunClient.jar and place the shortcut on your desktop.
If the Logon window does not come up, open command window, navigate to the client directory, and enter
java -jar RunClient.jar to see the output. This information will help you or Applications Manager support
troubleshoot the problem.
When starting the local client by double clicking on RunClient.jar (or a shortcut to it) in version 9.3.1 or above,
if it fails to start the error will now be in the startClient.log file.
10. On the Logon window, specify your User, password, automation engine.
User names are not case sensitive, but passwords are.
If you do not know your User name and password, check with your Applications Manager administrator.
11. If the Language field is available, select your language.
Specifying languages for the client is a desupported feature. The Language field is only available in
Applications Manager systems where languages were translated in a previous version.
12. On the Logon window, specify the following:
• The default user name SQLOPER.
• The default password is s0pass, where "0" is zero.
13. To have Applications Manager remember your User name and Automation Engine, select the Remember
logon button.
14. To accept the entered information and log on to Applications Manager, click OK.
If you have a large number of output files or tasks in the Backlog, Applications Manager will display a message
alerting you that the display has been truncated.
When you log out of Applications Manager, the following settings are remembered the next time you log back
in:
• The size and position of Explorer and Backlog Gantt View windows.
• The size and position of the main Applications Manager client window.
• The content pane of the Explorer window (if open).
• Whether a saved Backlog/History Filter was selected.
Upgrading the Client
To upgrade to a new version of the client:
1. Download the new version of the client Client.zip from the URL
set during the installation
your Applications Manager sent you
.
2. Delete everything except the connections.properties and client.properties files from your client directory.
3. Put the files from the new version of the Client.zip file into your client directory.
Receiving RmiServer Error Broadcasts
If the Receive RmiServer Error Broadcasts User Option is assigned to your User, the RmiServer Errors window
might pop up with one or more errors when you log in.
Applications Manager 9.4.1
When the Receive RmiServer Error Broadcasts User Option is assigned to a User, any RMI errors will be
displayed in a pop-up window as they occur. Additionally, if any RMI errors occurred since the last time the RMI
server was started, the last ten errors will be displayed in a pop-up window each time the User logs on.
If you have the DBA User Group, you can clear the RMI errors without stopping the RMI server by selecting Clear
RMI Errors from the View menu on the Applications Manager Desktop.
Re-Logging on to Applications Manager
There may be times when you want to log on to Applications Manager under a different user name or connect to
a different Automation Engine. To change your logon from the desktop, go to the File menu and select Re-Login.
You will have to re-enter your password when you re-log in, unless the Disable clearing of login passwords on
re-login Automation Engine option is checked.
Changing Your Password
You can change your password at any time. Your Applications Manager administrator can also set your password
to expire after a certain period of time.
To change your Applications Manager password go to the Options menu on the Applications Manager desktop and
click Change Password.
Accessing the Client Through a Firewall
If you will be accessing the Applications Manager client through a corporate firewall, you must open the appropriate
ports, and specify those ports in the Options.properties file on the host machine.
For details, see Configuring the Applications Manager Client.
For details, see the Installation Guide.
You can also request a Job or Process Flow from it's definition. This opens the Submit window, like the Request
window does. This is beneficial for testing changes you make to the object.
Procedure
To request one or more Jobs from the Requests window:
1. Open the Requests window shown above by doing one of the following:
• Open the Activities menu and select Requests.
• Select the Requests icon from the toolbar.
2. If appropriate, select an Application from the Application list box on the left side of the screen.
Applications specify a group of Jobs and Process Flows. The Application you select determines the Jobs and
Process Flows displayed in the table. Only the Applications and Jobs assigned to you via User Groups will be
displayed.
3. Select one or more Jobs from the list of Jobs and Process Flows on the right side of the screen. To select more
than one object, use Shift-Click or Ctrl-Click.
When selecting a Job/Process Flow, you can type the first few letters of its name in the Search field and
Applications Manager will find it. The Search field accepts valid regular expressions.
You can double-click to quickly request a single Job or Process Flow.
4. If the Requestor field is active, you can select the User that will be assigned to the Job/Process Flow.
You must have the Select Requestors User Option assigned to you by your Applications Manager administrator
to select a different User in this field.
5. To have the Requests window automatically close upon request, check the Close on Request box.
6. To request the Job(s), click Request.
Applications Manager opens the Submit window and displays the selected items as tabs. Each Job can
be viewed by selecting its tab. In the image below, three Jobs have been requested: JOB_REPORT,
MODULE_REPORT, and REPORT_BATCH.
Applications Manager 9.4.1
If any Jobs or Process Flows you select have Agent Groups assigned to them with application Agents, you will
be prompted in a pop-up window to select which Agent you want to use to validate prompts. You will still be able
to select the Agent or Agent Group for these tasks to run on from the Submit window.
Next Step
After requesting one or more Jobs and/or Process Flows, the next step is to enter prompt values, select options,
and submit. For more information, see Submitting Jobs and Process Flows.
Procedure
To respond to prompts and submit options:
1. Respond to the prompts in the Prompts box. Prompts are defined by an Applications Manager developer and
are specific to the Job or Process Flow. They are most often used to pass values to the program being run
by a Job. When a prompt includes a Select button as shown below, you can click it to select a single value or
multiple values from a list of options (depending on the Job/Process Flow definition).
Selecting values from lists helps to eliminate data entry errors. For information on selecting options from a list,
see Working in the Applications Manager Windows.
2. If the Send To option is active, select an Output Device from the drop-down list.
The Output Devices displayed in the list are determined by the Output Group assigned to the Job or Process
Flow. If the Send To option is not active, an Output Group was not assigned to the Job/Process Flow. You can
still run the Job/Process Flow and view output online, but the output will not be sent to an Output Device.
3. If the Send Option is active, select an option from the drop-down list.
The Send Option will be active if an output option is defined for the Output Device you select.
4. Select an output function from the Output Function drop-down box The output function determines how
output is handled. With any of these settings, the application output or report files and the system output files
are viewable from the Explorer window. There are three choices:
• LOG: Legacy setting that should not be used unless you need to use the Output window rather than the
Explorer window. For Jobs that have the output function set to LOG, Applications Manager loads all output
Applications Manager 9.4.1
or report files in the Output window every time you log into the client. This can take several seconds or
minutes. If more than 500 files are loaded, an alert will be displayed. Therefore, if you are not using the
Output window (that is, you view output from History instead of the Output window), you should use the
STORE setting.
• PRINT: The output is printed according to the output settings specified on this tab.
• STORE: The output is not printed.
5. If you are printing and want more than one copy of the output generated by the task, enter a value in the
Copies field.
6. Select a Queue from the Queue list box
The Queue list box will be active only if you have been assigned the Select Queues For Requests User
Option by your Applications Manager administrator. If you have not been assigned this option, the Queue list
box will be read-only, and will list only the default Queue of the Job.
7. If you are requesting a Job that was assigned to an Agent Group (or a Process Flow assigned to a multi-
execution Agent Group), you will be able to select a specific Agent from the Agent drop-down list.
8. The Requestor field is read-only and shows the user name that is assigned to the task. This name will be
displayed in the Requestor columns in the Backlog and History.
9. To set a different start date and time, place the mouse cursor in the Start Date field and enter a new date or
time. Or, click the button to the right of the field to open a window where you can select the date and time.
If you set the date and time forward, the task status in the Backlog will be shown as DATE PENDING.
Components of a Process Flow in DATE PENDING status will show as STAGED or STAGED_PW status
(depending on the components' predecessor requirements). If you set the date or time to a time that has
already passed, the task will be eligible to run immediately.
You can pre-date the start date of this Job or Process Flow to meet the predecessor requirements for tasks
waiting in the Backlog from a previous virtual day.
10. If you want the task to go into a HOLD status when it is submitted to the Backlog, you can select the Hold
option. The task will stay in the Backlog with a HOLD status and not run until you reset it.
11. If documentation has been created for the Job, you can view it by clicking the Documentation button.
12. When you have entered the prompt values and set the options, you are ready to click one of the following to
submit the Job or Process Flow:
• To automatically close the Submit window after it the Job is submitted, click the Submit & Close button.
• To keep the Submit window open, click the Submit button. After you submit the Job or Process Flow this
way, Applications Manager displays a message in the status bar (see below). If you submit a large Process
Flow the message will read, 'Task submission in progress: run_id = <run ID number>' until all components
of the Process Flow have been placed into the Backlog. The Close button will be grayed out until the task
is in the Backlog. Once the task is in the Backlog the message will read, 'Task was Successfully submitted
run_id = <run ID number>'.
Applications Manager 9.4.1
Removing Tabs for Jobs and Process Flows from the Submit Window
If there is more than one tab in the Submit window, you can remove the tab by selecting it and clicking Close Tab.
To close the Submit window and all tabs, click Close.
Invalid Prompts
After you click Submit, Applications Manager checks the prompt values. If the values you assigned to prompts are
invalid, Applications Manager will display an error message. You can correct the values and then submit the Job or
Process Flow.
Adding a Suffix to a Job or Process Flow Name So Predecessor Links Are Not Satisfied
To add a suffix to a Job or Process Flow name, enter the suffix in the Task Name Suffix field. The Job or Process
Flow will run with an underscore followed by the suffix at the end of its name. This will prevent the request from
satisfying any predecessors, since the name is changed.
The Task Name Suffix field will only be available if you have been assigned the Set Task Name Suffixes User
Option by your Applications Manager administrator.
task. The output generated by the task may exist in the out directory where the Automation Engine is installed, or
in a directory on another server associated with the application. Where output is registered is determined by the
Program Type script associated with the Job.
After you submit a task and it executes, you can view the output as plain text, HTML, or rich text format using the
File Viewer window shown above. You can also use an alternate viewer using file associations. For information
on setting file associations, see Opening Output Files with Other Applications . Output can be printed to a local or
system Output Device.
Opening the File Viewer Window
You can access the File Viewer window from:
• The Explorer window.
• The Output window.
The Explorer window is used to monitor and manage Applications Manager tasks, as well as Agents and Queues.
To open the File Viewer window from the Explorer window, right-click a task in the Backlog or History and select
Output from the pop-up menu. For more information on viewing output from the Explorer window, see Task Output
Files: Viewing.
The Output window provides access to task output for Users who have not been given access to the Explorer
window. To open the File Viewer window from the Output window, select a task and click the View button. For
more information on viewing output from the Output window, see Working with the Output Window .
Querying for Tasks
Applications Manager 9.4.1
You can query History in the Explorer window and you can query the Output window. You can search for specific
tasks by criteria such as Job name, Process Flow, Agent, and requestor. For more information on querying, see
topics Querying for Tasks in History and Querying the Output Window .
Viewing Output in Other Applications
After a task has completed, you can view the output in the File Viewer window as shown above. You can
also associate types of files with other viewers. For example, if you are generating an .xls file, you can have
Applications Manager automatically launch Microsoft Excel as the viewer. To do this, you must specify the
association in the File Association window. For information on setting file associations, see Opening Output Files
with Other Applications .
Printing Output
After viewing a report, you can preview the printed output and print it to a local Windows printer or to an
Applications Manager Output Device. These options are available from the File Viewer window's File menu and
from the icons in the File Viewer and Output windows. For more information on printing output, see Printing,
FTPing, and Emailing Output Files.
When you are setting up User Groups, consider setting up a User Group for end-users that includes access to the
Requests and the Output windows only. This lets end-users submit, view, and print their specific task requests.
Working with the Output Window
The table below describes the available actions on the Output window.
To: Do this:
View output Select a task and click the View button. Applications
Manager displays the output in the File Viewer window.
For information on using the File Viewer window, see
Viewing Output Files with the File Viewer.
Query for specific tasks Click Query on the Output window. You can query
for a specific task by a wide range of criteria such as
Job name, Process Flow name, requestor, and output
status. For more information on querying the Output
window , see Querying the Output Window .
Hide or unhide tasks Select one or more files and click the Hide button. To
select multiple adjacent files, use Shift+click. To select
multiple non-adjacent files, use Ctrl-Click. Hiding output
changes the status of the selected files to STORED and
removes the listing from the viewable list. To hide all
listings in this window, request the OLDPRTS Job.
To display hidden files, you must use the Query
function and query for the STORE status. If after
querying, you select a file and click Hide, its output
function will be changed to LOG.
If the Hide Files on the Output Window option was
not assigned to you by your Applications Manager
administrator, you will not be able to use the Hide
button.
Refresh the display Click the Refresh button. When you first open the
Applications Manager client, tasks that have completed
with a LOG status will be available for viewing without
running a query on the Output window. If other tasks
complete while you have a client session open, they
will not automatically be added to the Output window.
Refresh the display to see them.
Print output Select a task and click the Print button or one of the
Print icons. For more information on printing output, see
Printing, FTPing, and Emailing Output Files.
Procedure
To perform an output query:
1. Click the Query button on the Output window.
Applications Manager displays the Output query window shown above.
2. Select search criteria for your query.
You can use any combination of query criteria available in the Output query window.
The Output query window fields are described below.
• Jobs
Searches for any task assigned to the specified Job(s), including tasks that ran the Job under an alias name.
To query for a task run under an alias, enter the alias name.
• Process Flows
Searches for tasks that are part of the Process Flow(s). The Process Flows themselves are not returned.
• Agents
Searches for any task run on the specified Agent(s).
• Output Devices
Searches for any task assigned to the specified Output Device(s).
• Requestors
Searches for any task run by the specified requestor(s).
• Output Function
Applications Manager 9.4.1
Searches for tasks based on the output function set for the Job at run-time. There are three statuses:
• LOG: The output is available for printing, but has not been printed. Outputs with this status appear on the
Output window list by default.
• PRINT: Same as LOG, but output is sent to Output Devices as well.
• STORE: You must query to see output files with this status. System output files are typically stored.
• From start time/To start time
These fields accept date/time information. If the Current day option is selected, the From start time and To
start time fields will be inactive.
• Run ID
This field references the unique number (format: 1320.00) assigned to each task by Applications Manager.
When using this field, include the decimal value where appropriate. Decimal values at the end of Run IDs
indicate that tasks have been restarted. Every task has .01 added to its Run ID each time it is restarted.
• Current day
The default is to query by the current virtual day. If you save a query with this option checked, it will be
applied each time you run the saved query.
The date and time of the current day are saved as an offset from the current time. That means that if you
save a query for tasks run in the last two days, it will always run a query for the tasks in the last two days.
There is no way to save a query from a specific date. If you uncheck this option, the From start time and To
start time fields become active. You then can enter specific dates and times for the query.
To query by one or more objects, enter values in each object's field by doing one or both of the following:
• Type in names of the objects separated by commas.
• Select from a list of objects by clicking on the object icon at the end of the field.
3. You may select a default sort order for the query by selecting a Sort option.
The sort order can be overridden from the Output window by selecting a different column name.
4. You may also optionally save this output query to use later by entering a name in the Filter name drop-down
box The output query will be saved when you click OK.
Saved Output queries can be recalled from the Filter name drop-down box for History queries, Output queries,
and filters of the Backlog and History.
5. Click OK.
Applications Manager displays a small animated window as it processes the query. Once the query is
processed, Applications Manager displays the results in the Output window.
Using the Object Assignment Windows
When you click on an icon at the end of one of the applicable fields, Applications Manager displays the object
assignment window where you can pick one or more objects to use in an output query. The windows will list
only the objects to which you have User Group access. Use the arrow buttons to move objects between the
Unassigned and Assigned tables. For more information on assigning objects, see Working in the Applications
Manager Windows.
Applications Manager 9.4.1
You can enter page and line numbers for viewing, and bookmark pages. Output is automatically displayed in one of
three formats: plain text, HTML, or rich text format.
You can use the File Viewer window shown above or an alternate viewer. For information on viewing output with
alternate viewers, see Opening Output Files with Other Applications .
Viewing Output Files
To view output, open the File Viewer window by doing one of the following:
• From the Explorer window, right-click a task in the Backlog or History and select Output from the pop-up menu.
For more information on accessing the File Viewer window from Explorer, see Task Output Files: Viewing.
• From the Output window, select a task and click the View button. For more information on accessing the File
Viewer window from the Output window, see Working with the Output Window .
From the File Viewer window, you can:
• Copy text to the clipboard with Ctrl-C.
• Go to the start or end of the output file using the Home and End keys when the output text is in focus.
• Scroll through the output by using the horizontal and vertical scroll bars.
• Close the window by pressing the Esc key.
Printing, FTPing, and Emailing Output Files
You can print, FTP, or email output from the File Viewer window using the print icons or from the Print menu on
those windows. For more information, see Printing, FTPing, and Emailing Output Files.
Applications Manager 9.4.1
2. Enter the text you want to find and click Find Next.
Check the Match Case checkbox for case sensitive searches. Applications Manager searches run from the
current location to the end of the file.
Bookmarking Pages
As you view a file, you can bookmark pages for printing by selecting the Mark Page button. Your marked page
numbers will display in the Pages box located at the bottom of the screen. When you print the output, only the
marked pages will be printed.
Changing Output Formats and Text Size
Applications Manager automatically selects HTML or rich text format styles based on each file's extension (.htm,
.html, .rtf). Other files default to plain text view. When using the plain text view, you can adjust the size of the text
used in the display by selecting a value from the Size list box at the bottom of the viewer. This changes the size
of the text in the viewer, but it does not impact the size of the text used when the report is printed. Output can be
viewed with an alternate viewer if a file association has been defined. For information on setting file associations,
see Opening Output Files with Other Applications .
Viewing the End of Files
To view the end of a text file, go to the Options menu on File Viewer window and select Tail. The tailing option
enables operators to observe the most recent end of file every 10 seconds by showing the report as it is being
printed to standard output. Use this function, similar to the UNIX tail -f <filename> command, when trying to
diagnose problems.
1. From the File Viewer, go to the Print menu and select Send To, or click the Send To icon.
Applications Manager displays the Choose an Output Device window shown above.
2. Select an Output Device, copies, and output option and click OK.
Applications Manager prints to the Output Device you selected.
If you have bookmarked one or more pages using the Mark Page button, Applications Manager will only print
the marked pages.
To use the system print option, you must have User Group access to the Output Device you wish to print to and at
least one Output Group it is assigned to. If you do not think you have the necessary User Group access, see your
Applications Manager administrator.
Previewing a Print Task
To preview a print task, go to the File menu and select Print Preview, or click the Print Preview icon.
FTPing an Output File
You can send an output file to your client machine or a network location using the FTP function. To FTP a file:
1. From the Actions menu, select FTP or click the FTP icon.
Applications Manager 9.4.1
Separate multiple email addresses with a space or semicolon. To select from emails assigned to Users, click the
Select button. Applications Manager opens a window where you can select the email addresses. You can decide
whether to add the output file as an attachment and include additional text using the box at the bottom of the
screen.
You also have the option to send the output file as an attachment, include it in the message body, attach it as a
PDF file, or include it in an attached zip file.
In order to send emails, you must specify email settings for the Applications Manager Automation Engine/Local
Agent.
After you create a file association, files with corresponding names will always, sometimes, or never open with
the associated application when you view the output from the Applications Manager client, depending on the file
association definition. You will also have the option to open any output file with the application you created the file
association for.
The file association(s) you create are associated with your User on your PC. They are not global settings unless
you share file associations for all Users as descried in Sharing File Associations with All Applications Manager
Users.
Adding File Associations
To add an output file association:
1. On the desktop, go to the Options menu and select File Associations.
Applications Manager opens the File Association window.
2. Enter a file pattern in the Pattern field.
The most common pattern for file associations are file extensions (for example .xls).
The Pattern field accepts valid regular expressions. Note that regular expressions are case sensitive.
3. In the Application field, type the name of the executable file for the Application you wish to associate with the
file type, or use the ... button to browse for it on your PC.
4. Select an option from the Use field.
To: Select:
has been selected, and the Select Program window is open. From the Select Program window, you can select
the application you want to use to view the selected output file.
FileAssociations=N:/Automic/fileassoc.properties
FileAssociations=//rs62/ndrive/Automic/fileassoc.properties
Also, / should be used instead of \. Using backslashes will prevent properties from being saved.
The location for the Options.properties file is:
Applications Manager 9.4.1
UNIX:
$AW_HOME/web/classes
Windows:
%AW_HOME%\web\classes
2. Log into the Applications Manager client and set up a file association.
Applications Manager will create the file you specified in step 1 in the directory you specified and update it with
the file association information. The file association will now be globally changed for all Users that have access
to the drive you specified.
is minimized, the button on the taskbar uses the same color scheme. For more information, see Monitoring with the
Status Bar and Object Icons.
Customizing Explorer Tables
You can choose the columns you want displayed in the Explorer window and customize their order and names. For
more information, see Customizing Tables .
Viewing Output Files and Task Details
You can view output files and other task details for any task in the Backlog or History. You view the task details for
a task by right-clicking it and selecting an option from the pop-up menu. For more information, see Viewing and
Editing Task Details .
Staging Tasks in the Backlog
Staging tasks in the Backlog allows you to edit task details before the tasks are scheduled to run. For more
information, see chapter Staging Tasks in the Backlog.
Monitoring and Managing Queues and Agents
Queues control the flow of tasks. All tasks must pass through a Applications Manager Queue to be executed.
Agents are instances of Applications Manager; an Agent is installed on each machine where tasks are executed.
An Agent can be an Automation Engine's Local Agent, or a Remote Agent. The Automation Engine is also listed
along with the Agents in the Explorer window. You monitor and manage Agents and Queues from the Explorer
window. For more information, see chapter Working with Agents and Queues.
Monitoring and Managing Tasks in the Gantt View
You can monitor and manage tasks in the Backlog using the Gantt View. For information on using the Gantt View,
see chapter Monitoring and Managing Tasks with the Gantt View.
Explorer Terms
The following Explorer terms are used in this guide:
• Object tree: The graphical model displayed on the left pane of the Explorer window.
• Object icons: The icons used in the object tree to represent objects such as tasks, Process Flows, and
Queues.
• Object keys: The icons to the left of the expandable objects in the object tree. You can click the object keys to
show and hide objects in the tree.
Printing the Object Tree or a Table
You can preview and/or print the object tree, and the Backlog and History tables by selecting the appropriate
options from the File menu. The image below shows the print preview screen for a task in the Backlog in an
ABORTED status.
Numbers to the right of the label for the Backlog and History indicate the number of rows currently displayed in
each. Every row is counted regardless of whether it represents a Job, Process Flow, or historical record. The
Backlog is a list of tasks that:
• Are waiting to run.
• Are running.
• Have run and failed, and have stayed in the Backlog for operator intervention.
To view all tasks in the Backlog in the content pane on the top right side of the screen, click the Backlog icon in the
object tree. In the image above, one task is running in the Backlog, some tasks are waiting in a PRED WAIT status,
and one task has aborted.
Whether a task remains in the Backlog when it fails is determined by the Stay in queue on abort setting in its Job
definition.
Tasks sometimes fail by aborting, timing out, being killed, etc. When a task fails, a record is written to History under
the current run ID number. If a failed task remains in the Backlog, its run ID is incremented by .01.
For example, in the image above, a task with the run ID 73016 aborted and remained in the Backlog. A record of
the task aborting is written to History, and the task remains in the Backlog with the run ID 73016.01.
From the Backlog, you can:
• Take actions on tasks (see Taking Actions on Tasks in the Backlog).
Applications Manager 9.4.1
• View and edit task details (see Viewing and Editing Task Details ).
When a task leaves the Backlog, Applications Manager writes a record for it in History.
Numbers to the right of the label for the Backlog and History indicate the number of rows currently displayed in
each. Every row is counted regardless of whether it represents a Job, Process Flow, or historical record. Some
tasks may include multiple records in History. For example, a task that aborts and is reset will include two History
records.
History is an audit trail that includes records for:
• All completed tasks and occurrences of task failure(s).
• RMI servers or Agents being started, stopped, or having errors.
• The AgentService processes being stopped.
How long task history records are archived in the Applications Manager database is determined by the prompt
setting for the HISTORY_PURGE Job, which is part of the SYSTEM Process Flow. The default value is 60 days.
Your Applications Manager administrator is responsible for setting this value.
Applications Manager 9.4.1
The Explorer window includes a partial view of History in its lower right pane.
The amount of History displayed when any User logs in to the client is determined by the HistoryRetentionTime
setting in the Options.properties file.
How long records are displayed in the History pane while a User is logged into an Applications Manager session is
determined by:
• Your History Display Minutes desktop setting. For more information, see Editing General Desktop and ToolBar
Settings.
• Whether you have a History query activated. For more information, see Querying for Tasks in History.
From History, you can:
• Unsatisfy tasks as predecessors (see Unsatisfying Tasks as External Predecessors in History).
• View task details (see Viewing and Editing Task Details).
• Add comments (see Task Comments: Adding and Viewing).
• View History Gantt Views (see Viewing History in a Gantt Chart).
• Re-request tasks as described below.
Re-Requesting Tasks from History
You can now re-request a single task listed in the History pane of the Explorer window. To do so, right-click the
entry and select the Request option. Applications Manager lists the selected Job or Process Flow on a tab on the
Submit window.
When you re-request Process Flow components this way, they will run with the same prompt values as the selected
entry in History. All other settings, including alias name, conditions, predecessors, output settings, and Queues will
use the settings in the Job definition.
The default columns in the Backlog and History are described below. Columns unique to either the Backlog or
History are noted.
• Queue
The Queue the task is running on, will run on, or has run on.
• Run ID
The unique identification number Applications Manager assigned to the task.
Decimal values at the end of Run IDs indicate the number of times a task has either been killed or has aborted
and stayed in the Backlog. Each time a task is killed or aborts and stays in the Backlog, .01 is added to the
task's Run ID, and a record is added to History using the task's previous run ID.
• C
Uses a Y to identify tasks that include comments.
• D
Uses a Y to identify tasks that include documentation.
• Task Name
The name or alias, if defined, of the task.
• Start Date
Applications Manager 9.4.1
Your Applications Manager administrator can set default values for tables. Therefore, the columns in the Backlog
and History may be different from what is described above. For more information, see Customizing Tables.
The icon you select in the object tree determines the tasks listed in the Backlog.
The status bar reflects the most severe status in the system. The status bar colors and descriptions are described
below.
Color Description
For a description of the Automation Engine/Agent status values, see Automation Engine/Agent Status Values . For
a description of the task status values, see Task Status Values .
The status bar displays the time that the Explorer window was last refreshed. You can manually refresh by
entering Ctrl-R or by clicking the status bar on the Explorer window. The Explorer window is automatically
refreshed based on the Explorer Refresh Seconds desktop setting. For information on editing desktop settings, see
Editing General Desktop and ToolBar Settings.
The current date and time of the database are displayed to the right of the status bar.
Viewing Components in a Process Flow
Process Flow components in the Explorer tree are listed based on the execution order of each tasks' predecessors.
When you select a Process Flow in the Explorer tree, those same components are displayed in the top right
Explorer pane according to your Backlog search criteria (by default this is based on task status). To view the
structure of a Process Flow, right-click and select Flow Diagram.
Managing Task Statuses with the Object Icons
The Explorer icons in the object tree alert you to task status and give you a quick method for finding aborted/on
hold tasks.
For example, PROCESS FLOW_1 is in the Backlog. PROCESS FLOW_1 includes four Process Flow components.
The first three icons represent Process Flows that ran successfully. When the fourth Process Flow ran, a task
aborted, leading to the sequence of events below.
Applications Manager 9.4.1
1. When the task aborted, the status bar turned red to alert the user Pat Brown.
2. Pat opened the Explorer window and could see by the red X indicator on the Process Flows icon that one or
more tasks in a Process Flow had aborted.
Pat could have selected the Backlog, Ad Hoc, or Status icons to find the aborted task, but chose the Process
Flow option out of personal preference.
3. Next, Pat clicked the Process Flows icon key to view a list of running components.
The icon for PROCESS FLOW_1 had a red X indicator, so Pat knew that this Process Flow, or a Process Flow
nested therein, contained an aborted task.
4. Pat clicked the PROCESS FLOW_1 icon key to view the list of tasks belonging to PROCESS FLOW_1.
Pat could see that three Process Flows had finished successfully because the Process Flow icons were gray,
and that a task had aborted in the other Process Flow (because the Process Flow icon had a red X indicator).
Applications Manager 9.4.1
Pat clicked the PROJECTED_NOV icon key to view the components in that Process Flow.
Pat could see by the red X symbol as the second task's icon showing that it aborted. A second listing for the
task, an orange icon with a circle and a diagonal line notes that a history record of the task aborting was written.
Now Pat could view the task details including the output files for the task to see why it aborted, and reset or delete
it. For information on troubleshooting failed tasks, see Troubleshooting Failed Tasks .
The system output file includes basic information about the task such as:
• The name of the program executed.
• The person who ran the task.
• The Applications Manager run ID assigned to the task.
• The parameters used to run the task.
• A record of the basic steps taken to execute the task.
• Standard error reporting.
Getting More Information with Task Level Debug
If the system output file does not provide enough information to solve your problem, you can turn on the task level
debug. When task level debug is on, additional information is written to the task's system output file. If you call
Braodcom Support to help figure out why a task failed, they will most likely ask you to turn on task level debug,
reproduce the problem, and send in the log. Doing this ahead of time will speed you along to the quickest possible
resolution.
To turn on task level debug, create an empty file named BODY in the debug directory of the Automation Engine.
Action When the Process Flow is not When the Process Flow is
INITIATED INITIATED
Hold Holds the Process Flow header. Holds all eligible components in the
Does not affect its components. Process Flow.
Applications Manager 9.4.1
Action When the Process Flow is not When the Process Flow is
INITIATED INITIATED
Kill A Kill cannot be taken against Kills all eligible components in the
a Process Flow that is not in an Process Flow.
INITIATED status.
Reset Resets the Process Flow header. Resets all eligible components in the
Does not affect its components. Process Flow.
Delete Deletes the Process Flow header Deletes all eligible components in
and all its components.* the Process Flow.*
Remove all predecessors Removes all predecessors for the Removes all predecessors for the
Process Flow header. Does not Process Flow header. Does not
affect its components. affect its components.
* When Process Flows are deleted, skipped, or canceled the following will occur:
• All BEFORE conditions will be canceled.
• The Process Flow's children that are unstarted Process Flows will be initialized.
• All predecessors for the Process Flow's children will be removed.
Procedure
To put tasks on hold:
1. Select one or more tasks in the Backlog and right-click.
Applications Manager displays the pop-up menu shown above. The number to the right of the Hold, Kill, Reset,
and Delete options references the number of tasks you have selected that are eligible for that operation.
2. Select the Hold option.
Applications Manager displays the Hold confirmation window.
3. Click Yes to hold the task(s).
Applications Manager closes the confirmation window and displays a small animated window to show you that it
is taking the hold action on the task(s).
Result
When you put a task on hold, its status is displayed in a hold status, such as HOLD or HOLD PRED WT.
Applications Manager displays the status bar at the bottom of the window in yellow to alert you that one or more
Applications Manager 9.4.1
tasks are on hold. A record is written to the task's comment, noting that it was manually put on hold. The task
remains on hold until you reset the status or delete the task from the Backlog.
Killing Tasks
If a task is running, you can kill it by selecting the task and using the Kill command. When you kill a task, it stays
in the Backlog until you delete or reset it. When you kill a task, Applications Manager makes an entry in History
showing that the task was killed.
Applications Manager 9.4.1
Procedure
To kill running tasks:
1. Select one or more running tasks in the Backlog that you wish to kill and right-click.
Applications Manager displays the pop-up menu shown above. The number to the right of the Hold, Kill, Reset,
and Delete options references the number of tasks you have selected that are eligible for that operation.
2. Click the Kill option.
Applications Manager displays the Kill confirmation window.
3. Click Yes to kill the task(s).
Applications Manager closes the confirmation window and displays a small animated window to show you that
it is taking the kill action on the task(s). Once the task(s) go into a KILLED status, the Backlog label will be
displayed in its default color.
Result
While Applications Manager is killing a task, the task's status in the Backlog changes briefly to KILL. Once the task
is killed its status changes to KILLED, and an entry is made in History to show this (see image below). A record is
Applications Manager 9.4.1
written to the task's comment, noting that it was manually killed. The killed task stays in the Backlog until you delete
it or reset it.
When you kill a task in Applications Manager running in a UNIX environment, Applications Manager issues the
kill -15 UNIX command followed by the kill -9 UNIX command.
Procedure
To reset one or more tasks from the Backlog:
1. If appropriate, change the task details.
For more information on changing task details, see Viewing and Editing Task Details.
2. Select one or more tasks in the Backlog and right-click.
Applications Manager displays the pop-up menu shown above. The number to the right of the Hold, Kill, Reset,
and Delete options references the number of tasks you have selected that are eligible for that operation.
3. Select the Reset option.
Applications Manager displays the Reset confirmation window.
4. Click Yes to reset the task(s).
Applications Manager closes the confirmation window and displays a small animated window to show you that it
is taking the reset action on the task(s). Once the task(s) go into a LAUNCHED status, the Backlog label will be
displayed in its default color.
Result
Applications Manager 9.4.1
A record is written to the task's comment noting that it was manually reset. When a thread becomes available for
the Queue, Applications Manager will start the task(s).
Deleting Tasks
If a task in the Backlog is in a non-running status, you can delete it. For example, tasks with a status of
LAUNCHED, PRED WAIT, ABORTED or KILLED can be deleted. You can also delete Process Flows from the
Backlog. If you delete a Process Flow, all components in the Process Flow are deleted as well. When you have
deleted a task, you cannot reset it from the Explorer window.
Applications Manager 9.4.1
Procedure
To delete tasks:
1. Select one or more tasks in the Backlog and right-click.
Applications Manager displays the pop-up menu shown above. The number to the right of the Hold, Kill, Reset,
and Delete options references the number of tasks you have selected that are eligible for that operation.
2. Select the Delete option.
Applications Manager displays the Delete confirmation window.
3. Click Yes to delete the task(s).
Applications Manager closes the confirmation window and displays a small animated window to show you that it
is taking the delete action on the task(s). Once the task(s) go into a DELETED status, the Backlog label will be
displayed in its default color.
Result
Applications Manager 9.4.1
When you delete tasks, they are removed from the Backlog. A record of the task deleted from the Backlog is now
displayed in History with a DELETED status as shown below. An entry is also made in the task's comment, noting
that it was manually deleted. After deleting a task, you cannot reset it from the Explorer window.
To remove all predecessor links for a task in the Backlog, right-click the task and select Remove All Predecessors
from the pop-up menu as shown above. The task will then be eligible to run as shown below.
Applications Manager 9.4.1
If you wish to view and remove select predecessors for a task, you can do so on the Predecessors tab of the Task
Details window. For more information, see Pending Predecessor Links: Viewing and Removing.
When you remove predecessor links for a task in the Backlog, the changes are applied only to that instance of the
task. They do not affect the definition of the Job or Process Flow.
Removing External Predecessors by Queue
To remove a task's external predecessors on insert to the Backlog, assign it to the
REMOVE_EXT_PREDS_ON_INSERT Queue. You can create the REMOVE_EXT_PREDS_ON_INSERT Queue
by hand or by importing it. For more information, see Removing External Predecessors by Queue.
• Documentation: Information written by the person who created the Job or Process Flow. Documentation
provides relevant information about the processing of a task. It can be comments, suggestions, or instructions.
Documentation cannot be altered from the Explorer window.
• Output Files: Output files created on the Agent machine when the task is run. Output files include system and
application files for the task. Output files can be opened for viewing using the File Viewer window.
• Comments: Information about the running of a task. Applications Manager automatically creates comments that:
• Tell about condition actions that affect the running of a task.
• Detail DELETE, HOLD, RESET, and KILL actions taken by a User.
• Give details on tasks with a LAUNCH ERR status.
Additionally, you can include your own comments to provide relevant information about the processing of a task.
Comments are not files, and should not be confused with documentation or output files.
• Run ID
A unique number assigned to a task by Applications Manager at the time the task runs.
• Process Flow Name
The parent Process Flow, if this task is a Process Flow component.
• Process Flow Order No.
The order number of the task in its Process Flow, if the task is a Process Flow component.
• Request Date
The actual time that the task was placed into the Backlog. Corresponds to the so_request_date database field.
• Started
The actual time that the task was started by the Automation Engine and processed by the Agent. Corresponds
to the so_job_started database field.
• Ended
The actual time the task finished processing on the Agent. Corresponds to the so_job_ended database field.
• Status
The task status. For a description of the task status values, see Task Status Values.
• Process ID
A unique number assigned to a task by the operating system.
• Status Detail
Details that give additional information on the reason for a task's status.
• Other Details
When clicked displays a pop-up window with the application status and reference fields. These are used with
the PeopleSoft extension and some custom solutions.
• Notification, Output Scan, Environment Variables
Lists the Notification, Output Scan, and Environment Variable objects assigned to the task and where they are
assigned.
Editable fields
• Max Run Time
Used to prevent runaway programs. It determines how long the program can run before timing out
(DDD:HH:MI). A time of 0 lets the task run forever. If a task times out, it is given a status of TIMEOUT.
• Priority
Determines when a Job is run in relation to other tasks in the same Queue. A task with a lower priority number
will be run before tasks with higher numbers. The range is 1-99. The default priority is 50.
Tasks submitted with 0 priority will have the status of NO PRIORITY in the Backlog and will not run.
The execution order of tasks waiting to run in a QUEUED status is decided in the following order:
1. Queue priority
2. Job priority
3. Start date and time
Therefore, if two tasks are waiting to run in different Queues, and those Queues have the same priority, the
Jobs' priorities are checked. If Queue and Job priorities are the same, their start date and times are compared.
• Login
The Login the system will use when the task is executed. Logins allow operators and programmers to run
programs that access a database or host without having to know the login and password.
Primary and secondary Logins can be specified for Jobs. The primary Login can be overridden (if one is defined
for the Job) by:
• Jobs requested with the REQUEST JOB condition action.
• Process Flow components.
• Queue
The Applications Manager Queue the task will run through. Queues are assigned to Jobs and Process Flows.
If a schedule is defined for a Job or Process Flow, and the schedule is assigned to a Queue, the schedule's
Applications Manager 9.4.1
Queue will override the Jobs or Process Flow's Queue. The Queue setting for a Process Flow does not affect
its components, unless the Insert components into Process Flow's Queue Automation Engine option is
turned on.
The Process Flow's Queue is used for components only when the Insert components into process flow's
queue Automation Engine option is turned on.
Queues can be specified for requests (as long as the Request Queues User Option is assigned).
• Agent
The Agent where the program is stored and run.
• Start Date
The time that the task was scheduled to start. Corresponds to the so_start_date database field. Start dates for
tasks in the Backlog are determined by either:
• Requests (may be post-dated).
• Schedules for Jobs and Process Flows.
• The schedule of a component's Process Flow.
• Send To
Specifies the Output Device or set of devices where the output will be sent (for example: EMAIL, ACCOUNTING
LASER, ATLANTA LASER).
• Output Option
Used for specifying dynamic output options for the selected Output Device. This value or list of values is defined
by the Output Interface assigned to the Output Device.
• Output Function
Determines how output is handled. There are three choices:
• LOG: Legacy setting, should not be used unless you need to use the Output window rather than the
Explorer window.
• PRINT: The output is printed.
• STORE: The output is not printed.
• Copies
Sets the default number of copies to be printed.
• Restart once on abort
When selected, Applications Manager will automatically restart a task the first time it aborts, but will not restart it
if it aborts a second time.
When this option and the Stay in Queue on abort option are both set, and the task aborts, you will see three
listings for it in the Explorer window.
• The original listing for the Run ID <run_id> in History shows that the task ran.
• A second listing <run_id>.01 in History shows that it aborted.
• The current listing <run_id>.02 in the Backlog represents the restarted task.
The Restart once on abort setting is specified in each task's definition, and can be overridden with a condition.
To view the task details for a task in the table, right-click the task and select an option.
Understanding Predecessor Link Types
Descriptions for each link type are given below:
• Started: Predecessor must have started or been skipped. Represented by a solid yellow line.
• Success since last run: For external predecessors only. Predecessor must complete with a status of
FINISHED since the last time this Job ran. Represented by a dashed blue line.
• Success (default): Predecessor must complete with a status of FINISHED or be removed from the Backlog.
Represented by a solid green line.
• Success only when FINISHED: For external predecessors only. Predecessor must complete with a status of
FINISHED. Represented by a solid blue line.
• Success (skip on failure): Predecessor must complete with a status of FINISHED. If status is ABORTED,
DIED, or TIMEDOUT, the component is skipped. Represented by a dashed green line.
Applications Manager 9.4.1
• Failure: Predecessor must complete with a status of ABORTED, DIED, or TIMEDOUT. Represented by a solid
red line.
• Failure (skip on success): Predecessor must complete with a status of ABORTED, DIED, or TIMEDOUT. If
status is FINISHED, the component is skipped. Represented by a dashed red line.
• Complete: Predecessor completes with any status including FINISHED, DIED, or ABORTED. Represented by
a solid black line. Predecessor links to a Process Flow must use this predecessor link type or the Success link
type.
Removing Predecessor Links
To remove predecessor links from either pane of the Pending Predecessors window, check the box in the
Remove column for the task and click the Apply button. Applications Manager deletes the predecessor link. If
there are no other predecessor links, the task should run. This has the same effect as deleting a predecessor link
from the referenced task's Flow Diagram tab.
Deleting predecessor links from this tab does not delete tasks from the Backlog.
If you wish to remove all predecessor links for a task in the Backlog, you can do so by right-clicking and
selecting the Remove All Predecessors option from the pop-up menu. For more information, see Removing All
Predecessors for Tasks.
When you remove predecessor links for a task in the Backlog, the changes are applied only to that instance of the
task. They do not affect the definition of the Job or Process Flow.
Removing External Predecessors by Queue
To remove a task's external predecessor links on insert to the Backlog, assign it to the
REMOVE_EXT_PREDS_ON_INSERT Queue. You can create the REMOVE_EXT_PREDS_ON_INSERT Queue
by hand or by exporting it. For more information, see Removing External Predecessors by Queue.
To view the task details for a task in the table, right-click the task and select an option.
From the Flow Diagram tab, you can monitor and manage objects in much the same way as from Explorer. You
can edit predecessor links as well as take actions on tasks such as delete, hold, restart, and kill.
Viewing Predecessor Links in a Flow Diagram
To view the predecessor links for a task in a flow diagram, right-click the task and select Flow Diagram from the
pop-up menu. Applications Manager displays the Task Details window with the Flow Diagram tab selected.
External references to scheduled tasks that are not yet in the Backlog will have a dashed border.
Adding and Editing Predecessor Links in a Flow Diagram
You can add, edit, or delete internal and external predecessor links for non-running tasks in the Backlog. An
internal predecessor link is a link to a component within the parent Process Flow. An external predecessor is a
link to a component outside the parent Process Flow. When you edit a link, you can change the link type, redirect
the link to another component, or delete the link. You can also add external predecessor links for these tasks. You
cannot add components to a Process Flow from the Backlog.
Applications Manager 9.4.1
When you add, edit, or delete predecessor links for a task in the Backlog, the changes are applied only to that
instance of the task. They do not affect the definition of the Job or Process Flow.
You can select the other tabs to view and edit the task's general task details, prompts, documentation, output, and
more.
Disabling Pop-Up Tables
You can disable the pop-up tables by unchecking the Show status info option under the Options menu. The
Show status info setting can be saved for your workstation by selecting Save Preferences from the File menu.
Taking Actions on Tasks and Viewing or Editing Task Details
You can right-click tasks in the Flow Diagram tab to take actions on them or to view or edit their task details the
same way you would in the Explorer window. For more information on taking actions on tasks, see Taking Actions
on Tasks in the Backlog. For more information on viewing and editing task details, see Viewing and Editing Task
Details.
To add, edit, delete, or copy a condition for a task in the Backlog, select the appropriate button. When you make
changes to conditions for a task in the Backlog, the changes are applied only to that instance of the task. They do
not affect the definition of the Job or Process Flow. You can select the other tabs to view or edit the task's general
task details, prompts, documentation, output, and more.
Where Conditions Are Defined
Conditions can be defined for Jobs, Process Flows, and Process Flow components. The conditions are not
included when the Job/Process Flow is assigned to a Process Flow unless the Use Job Conditions option is
checked for the component.
If the task has HTML documentation, you can uncheck the HTML box to view the source HTML code.
2. In the top portion of the window, select the type of documentation you wish to view.
Applications Manager 9.4.1
Applications Manager displays the documentation you selected in the bottom portion of the window.
You can select the other tabs to view and/or edit additional details for the task.
3. Click OK to close the window.
When Job definitions include Program Types that ship with Applications Manager, standard output and error
file names begin with an 'o'. Output files generated by the task begin with a 'b'. If custom Program Types are
assigned to your Jobs, their output files may be named differently.
Applications Manager 9.4.1
The list for Process Flows will include the output files of all its components.
2. To print an output file without opening the File Viewer window, select a task and click the Print button or one of
the Print icons. For more information on printing output files, see Printing, FTPing, and Emailing Output Files.
3. To view a file, select the file and click View.
Applications Manager displays the File Viewer window. For information on using the File Viewer window and
printing output files, see Viewing Output Files with the File Viewer.
You can select the other tabs to view and/or edit additional details for the task.
After you have added an comment entry, you cannot change it. It becomes a permanent part of the task's history.
Querying Comments for Other Tasks
To view comments for another task:
1. From the Task Details window, select the Comment Query tab.
The table on the top of the screen displays all previous comments for the selected Job or Process Flow.
To disallow predecessor links to this task, you must unsatisfy it as a predecessor. Once a task is unsatisfied as a
predecessor, it is as if it did not run. All predecessor links to it will not be satisfied until the Job, Process Flow, or
Process Flow component runs again.
Procedure
To unsatisfy a task as a predecessor for all potential predecessor links, right-click the task listing in History and
select Unsatisfy for Predecessors from the pop-up menu. Applications Manager changes the task's status to
UNSAT-FINISH and unsatisfies this running of the task as a predecessor for all predecessor links which reference
it, as shown below. The predecessor links of other tasks will now need to be satisfied by another instance of this
Job, Process Flow, or Process Flow component.
Applications Manager 9.4.1
Applications Manager opens the History Gantt view window shown below.
Applications Manager 9.4.1
If you select Process Flows or Process Flow components, Applications Manager displays all tasks in the
corresponding Process Flow(s).
For detailed information on the features available in all Gantt windows, see chapter Monitoring and Managing
Tasks with the Gantt View.
3. If you wish, you can right-click a task and select Flow Diagram to view its predecessors in a flowchart view.
Procedure
To compare run times of tasks in History:
1. Select two or more records for Jobs and/or Process Flows in History that you want to compare.
You may want to run a History query of the tasks you wish to compare. For information on running History
queries, see Querying for Tasks in History.
In the image above, two instances of the INV_STATUS Job are selected.
2. Right-click the History Gantt view option from the pop-up window.
Applications Manager opens the History Gantt view window shown below.
Applications Manager 9.4.1
3. In the History Gantt view window, select Set start times to midnight from the Actions menu.
4. Applications Manager displays all start times as midnight so you can compare run times as shown below.
Applications Manager 9.4.1
Date Pending If the scheduled start date is still in the future, the status
will be DATE PENDING. If the start date has passed,
Applications Manager will launch the task as soon as
possible after the system is brought back online.
Applications Manager 9.4.1
When a task is running AND the machine AND the network THEN
and the Agent running the Agent
Jobs and Process Flows scheduled to run during the down time will run once when the system is brought back up.
They will then return to their normal schedule. It does not matter how many times the Job or Process Flow was
scheduled to run during the down time. Each Job or Process Flow will run only one time before returning to its
normal schedule, unless a date was entered in the Reschedule from field.
If the Oracle Tables Are Lost
If the Applications Manager Oracle tables are lost as a result of the system failure, all status information will be
lost. You will need to restore the Oracle database and let Applications Manager resume processing based on the
schedule information restored by the backup.
You can type the first few letters of an task from the Backlog's name in the Search field, and Applications Manager
will find it. The Search field accepts valid UNIX regular expressions. For example, to search for all Jobs starting
with the letters A and T, you would enter [at] in the Search field.
What Is History?
History is an audit trail including records for:
• All completed tasks and occurrences of task failure(s).
• RMI servers or Agents being started, stopped, or having errors.
• AgentService processes being stopped.
How long task history records are archived in the Applications Manager database is determined by the prompt
setting for the HISTORY_PURGE Job, which is part of the SYSTEM Process Flow. The default value is 60 days.
Your Applications Manager administrator is responsible for setting this value.
The Explorer window includes a partial view of History in its lower right pane.
The amount of History displayed when any user logs in to the client is determined by the HistoryRetentionTime
setting in the Options.properties file.
How long records are displayed in the History pane while you are logged into a Applications Manager session is
determined by:
• Your History Display Minutes desktop setting. For more information, see Editing General Desktop and ToolBar
Settings.
• Whether you have a History query activated.
You select search criteria for your query. You can query by one or more options by entering values in as many fields
as you like. Fill in the fields by doing one or both of the following:
• Typing in values
• Selecting values from a list
You may select a default sort order for the query by selecting a Sort option at the end of a field. The sort order can
be overridden from the History pane by selecting a different column name.
When you are through defining the query, click OK. Applications Manager displays a small animated window as it
processes the query. Once the query is processed, Applications Manager displays the results in the History pane of
the Explorer window.
Field descriptions for the History Query window are described below. Directions for the two methods of filling in the
fields follow.
• Jobs
Searches for any task assigned to the specified Job(s).
To query for a task run under an alias, enter the alias name.
• Process Flows
Searches for Process Flow(s) and components.
• Applications
Searches for any task assigned to the specified Application(s).
• Agents
Applications Manager 9.4.1
Use the Other filters field to type in comma-separated names. This field allows for wildcards and negative filters as
described in the "Typing in Values" section above.
Any text entered in the Other filters field will be included in the appropriate field on the Select filters window.
Adding the Queried Results to History
By default Applications Manager limits the records in the History pane to the results of the query. If you would rather
add the queried results to the current History display, click the Add to Current History box on the History Query
window.
When you add the queried results to the current History display, the records will be added to the History pane and
the Apply Query box on the Explorer window will not be checked. The records added from the query will remain
in History for the number of minutes you have defined in your History Display Minutes desktop setting. To see them
again, you can run another query. For more information on the History Display Minutes desktop setting, see Editing
General Desktop and ToolBar Settings.
Saving a Query
You may also optionally save a History query to use later by entering a name in the Filter name drop-down box
The History query will be saved when you click OK.
Saved History queries can be recalled from the Filter name drop-down box for History queries, Output queries, and
filters of the Backlog and History.
You can delete a saved query by selecting it and clicking Delete.
Viewing Query Results
When you run a query, the Apply Query checkbox will be selected and the column headers will be displayed with a
light orange background as shown below.
Applications Manager 9.4.1
Applications Manager will not refresh the query's results. If you want to view updated query results, you must re-run
the query.
To view the unqueried History, uncheck the Apply Query box. You can view the queried results again by
rechecking the Apply Query box. If you select the History Query menu item again, Applications Manager returns
you to the Query Definition window. Applications Manager displays the search criteria you defined on your last
query. To run a new query, select new data and click OK.
You select search criteria for your filter. You can filter by one or more options by entering values in as many fields
as you like. Fill in these fields by doing one or both of the following:
• Typing in values
• Selecting values from a list
You may select a default sort order for the query by selecting a Sort option at the end of a field. The sort order can
be overridden from the Backlog or History panes by selecting a different column name.
When you are through defining the filter, click OK. Applications Manager displays a small animated window as
it processes the filter. Once the filter is processed, Applications Manager displays the results in the Backlog and
History panes of the Explorer window.
Field descriptions for the Select filter window are described below. Directions for the two methods of filling in the
fields follow.
• Jobs
Searches for tasks assigned to the specified Job(s), including tasks running the Job under an alias name.
To query for a task run under an alias, enter the alias name.
• Process Flows
Searches for the selected Process Flow(s) and their components.
• Applications
Searches for any task assigned to the specified Application(s).
• Agents
Applications Manager 9.4.1
Use the Other filters field to type in comma-separated names. This field allows for wildcards and negative filters as
described in the "Typing in Values" section above.
Any text entered in the Other filters field will be included in the appropriate field on the Select filters window.
Viewing Filter Results
Applications Manager runs the search and displays the filtered results. The Apply Filter checkbox will be selected
and the column headers will be displayed with a orange background as shown in the image below. To view the
unfiltered Backlog and History, uncheck the Apply Filter box. You can view the filtered results again by rechecking
the Apply Filter box. If you select the Filter Backlog & History menu item again, Applications Manager returns
you to the Select filters window. Applications Manager displays the search criteria you defined on your last filter.
To run a new filter, select new data and click OK. To select a saved filter, select it from the drop-down box between
the two checkboxes in the History pane.
Applications Manager 9.4.1
Staging Tasks
Applications Manager staging allows you to edit tasks before they run. You tell Applications Manager which tasks
to stage by specifying prompt values and running either the STAGING or STAGING_BY_SCHEDULE Job. These
Jobs can be scheduled or requested on an ad hoc basis. Jobs and Process Flows should not be edited, and you
should not run exports or imports while these Jobs are running.
Applications Manager 9.4.1
• For STAGING: individual Jobs and Process Flows with schedules from the Application(s) selected in the first
prompt.
• For STAGING_BY_SCHEDULE: schedules from Jobs and Process Flows with schedules from the
Application(s) selected in the first prompt.
If one or more Applications are selected, but no Process Flows or Jobs, then all Process Flows/Jobs of the
selected Applications will be staged. Any scheduled Process Flows or Jobs not staged will be inserted in the
Backlog at their regularly scheduled time.
• Include Hour/Minute Schedules
Determine whether you wish to include Process Flows/Jobs with Hours or Minutes selected in their schedule's
Units box.
• Hours ahead to be staged
Determine the number of hours ahead you want to stage tasks.
By default, this prompt has a maximum value of 48 hours. This is so no one accidentally stages out too far. If
you wish to stage tasks out longer, you must edit the prompts in the Job's definition.
If the Use virtual day for Process Flow component days of week option is selected for the Automation
Engine, tasks cannot be staged beyond 24 hours from the start of the virtual day.
Warning: Staging tens of thousands of tasks can affect Backlog performance.
When you stage a task, the value in its schedule's Next run date field will be updated to its next run.
Creating Multiple Schedules
You can create multiple schedules for these Jobs that use different prompt values. For example, you might create
the following schedules:
• WORKDAY: Runs Monday through Friday at 9:00 A.M. and lists all Process Flows and Jobs for 8 hours without
including hour/minute schedules.
• EVERY_HOUR: Runs every hour and lists all Process Flows and Jobs including hour/minute schedules.
For a description of all task status values, see Task Status Values.
You can filter staged tasks in the Backlog using the Future Hours field for a Backlog & History filter. For more
information, see Filtering the Backlog and History.
Editing Staged Tasks
Staging tasks in the Backlog gives you the opportunity to alter a task's details before it runs. Editing staged tasks is
exactly the same as editing any other non-running task in the Backlog. For information on altering task details, see
Viewing and Editing Task Details.
Queues control the flow of tasks. All tasks must pass through an Applications Manager Queue to be executed. You
control Queue throughput by assigning a Queue to a Thread Schedule. You can define an unlimited number of
Queues.
Agents are instances of Applications Manager; an Agent is installed on each machine where tasks are executed.
An Agent can be an Automation Engine's Local Agent, or a Remote Agent. The Automation Engine is also listed
along with the Agents in the Explorer window. The Automation Engine schedules and controls task execution on all
the Agents assigned to it.
You monitor and manage Agents and Queues from the Explorer window.
Monitoring and Managing Agents
From the Explorer window, you can:
• View information in the Agent Summary including the status of the Automation Engine and Agent(s). An Agent
Summary is shown below.
Monitoring Agents
You can view an Agent Summary by clicking the Agents icon from the object tree.
Agents are installed on each machine where tasks are executed.
Applications Manager 9.4.1
Each Automation Engine has one Local Agent and can control numerous Remote Agents and application Agents.
Several Agents are defined and the Remote Agent TRUNK is selected in the image above. From the Explorer
window you can view:
• An Agent summary by clicking the Agents icon in the object tree.
• All Agents on a machine by clicking the machine name under the Machines icon in the object tree.
• Agent status based on icons in the Explorer tree and descriptions in the Agent Summary.
Column Descriptions for the Agent Summary
The Agent Summary displays the status of each Agent. The columns are described below.
• Agent
Name of the Agent, Agent Group, or Automation Engine.
• Status
Applications Manager 9.4.1
Current status of the Agent. For information on possible status values, see Automation Engine/Agent Status
Values .
• Type
Designates each listing as an Automation Engine or Agent type.
Although the Automation Engine and its Local Agent are defined as a single object in Applications Manager,
they are listed separately in the Agent Summary, and separate actions can be taken on them.
• Elapsed
Elapsed time of the Agent status.
• CPU
Percentage of CPU usage on the Agent machine. The CPU usage is updated about once every minute for each
Agent and the Automation Engine. CPU percentages are rounded to the nearest five percent.
• Max Tasks
Number of threads available for the Agent.
• Bklg
Number of tasks in the Backlog.
• Run
Number of tasks RUNNING.
• Hold
Number of tasks on HOLD.
• Abtd
Number of tasks ABORTED in the Backlog.
• Process Flow
Number of Process Flows in the Backlog.
You can customize many tables in Applications Manager, including the Agent Summary. When customizing tables,
you determine:
• Which columns are listed.
• What each column is named.
• How each column is displayed.
Your Applications Manager administrator can set default values for tables. Therefore, the columns in the Agent
Summary may be different from what is described above. For more information, see Customizing Tables .
Managing Agents
You manage the Automation Engine and its Agents in several ways. From the Explorer window you can:
• Take an action on a single Agent or Automation Engine by right-clicking an Agent's icon in the object tree or by
right-clicking a listing in the Agent Summary.
• Take an action on one or more Agents or the Automation Engine by highlighting them in the Agent Summary
and right-clicking.
• Take an action on all Agents and the Automation Engine by right-clicking the Agents icon in the object tree.
• Take an action on all Agents on a machine by right-clicking the machine name under the Machines icon in the
object tree.
• Change an Agent or Automation Engine's Thread Schedule by right-clicking the Agent or Automation Engine.
Actions you can take on Agents are start, stop, idle, resume, and reset. These actions can be taken on the
Automation Engine as well. Each action is described below.
Applications Manager 9.4.1
To manage Agents, you must have the Take Actions on Agents User Option assigned to you by your Applications
Manager administrator.
Idling and Resuming Agents or the Automation Engine
If you want to stop processing newly submitted tasks through an Agent you can idle it. The Agent will go to an Idled
status and the icon for the Agent will be displayed with a yellow triangle over it in the Explorer tree. Tasks in the
Backlog set to run on an Agent in an Idled status will have a task status of AGENT WAIT. To take the Agent out of
the idle status, you resume it.
Starting Agents or the Automation Engine
Starting an Agent will start processes on that Agent if they are stopped and change the Agent's status to Running.
The Agent may change to an interim Starting status before it moves to Running. Starting an Agent is equivalent to
issuing the startso <Agent name> command. It may take some time to update the status when starting Agents or
the Automation Engine.
Applications Manager 9.4.1
Thread Schedules are assigned to Automation Engines, Agents, and Queues. You can also specify a local or
Remote Agent's Thread Schedule in its definition. For information on defining Thread Schedules, see Defining
Thread Schedules.
The log will display in the File Viewer window. For detailed information using the File Viewer window, see
Viewing Output Files with the File Viewer.
Printing, FTPing, and Emailing Logs
To print, FTP, or email a log file without opening the file viewer, select a task and select an option from the Print
menu or one of the Print icons. For more information on printing output, see Printing, FTPing, and Emailing Output
Files.
Zipping Log Files
You can zip log files to compress them by selecting the file and clicking the Zip button. When the file is zipped, you
will still be able to view it from the Applications ManagerFile Viewer window.
You cannot zip current log files that are still receiving information.
Rolling Over AgentService.log and RmiServer.log Files
You can roll over the current file to create new AgentService.log files for Agents or the RmiServer.log files for the
Automation Engine. To rollover one of these logs, right-click on an Agent or Automation Engine and select Agent
Log Rollover or RmiServer Log Rollover. You might want to rollover a log file when debug is turned on and you
want to isolate what is written in a log.
Customizing the Agent Logs Table
You can customize many tables in Applications Manager, including the table in the Agent logs window. When
customizing tables, you determine:
• Which columns are listed.
• What each column is named.
• How each column is displayed.
Your Applications Manager administrator can set default values for tables. Therefore, the columns in the Agent
logs window may be different from what is shown above. For more information, see Customizing Tables.
Monitoring Queues
You can view the Queue Summary in Explorer.
Controlling the load on your system is critical. In Applications Manager, you control the workload by setting the
number of concurrent tasks that can pass through the Queues.
Applications Manager 9.4.1
Displays whether the Queue is active or inactive. Active Queues apply the thread settings. When you inactivate
a Queue, no tasks will be processed through it. The tasks will remain in the Backlog with a QUEUE WAIT status
until the Queue is activated.
The object tree icons for inactive Queues will be covered by a yellow triangle.
• Pri
The priority assigned to the Queue. Priority defines the order in which the Queue is scanned for task initiation.
Tasks in a higher priority Queue will be processed before tasks in a lower priority Queue. The lower the number
you assign to a Queue, the higher priority it will have to run tasks. If all threads in a high priority Queue are
being used, Applications Manager will process tasks in lower priority Queues until the maximum number
of threads is reached. You can control the load on your system and the availability of computing resources
by giving careful thought to how you prioritize Queues and set their thread limits. For information on Queue
priorities, see Administering Queues .
• Threads
Number of threads assigned to the Thread Schedule displayed in the Schedule column.
• Bklg
The number of tasks in the Backlog waiting to run on the Queue.
• Schedule
The Thread Schedule assigned to a Queue. Thread Schedules control the maximum number of concurrent
tasks that can run through the Queue at any given time. For information on defining Queues and Thread
Schedules, see Administering Queues .
You can customize many tables in Applications Manager, including the Agent Summary. When customizing tables,
you determine:
• Which columns are listed.
• What each column is named.
• How each column is displayed.
Your Applications Manager administrator can set default values for tables. Therefore, the columns in the Queue
Summary may be different from what is described above. For more information, see Customizing Tables.
Active Queues apply the thread settings. When you inactivate a Queue, no tasks will be processed through it.
The tasks will remain in the Backlog with a QUEUE WAIT status until the Queue is activated.
You can also activate or inactivate one or more Queues from the Queue Summary by highlighting the Queues,
right-clicking, and selecting the appropriate option. When you change Queue settings from the Queue Summary,
the Queue definition is altered in the database.
Applications Manager User Groups allow you to edit Queues. If you cannot edit them, see your Applications
Manager administrator.
Administering Queues
You control the flow of tasks to servers by using Applications Manager Queues. All tasks pass through an
Applications Manager Queue to get to a server. You control Queue throughput by assigning each Queue to a
Thread Schedule. You can define an unlimited number and type of Queues.
Thread Schedules control the number of concurrent tasks that can run through a Queue. When you define a
Thread Schedule, you specify the number of threads, the minimum threads, and the start and stop times for the
schedule. A Thread Schedule can be divided into sub schedules, letting you change the number of threads for
different times of the day.
For example, you might assign only one thread from midnight to 6:00 A.M., four threads from 6:00 A.M. to 5:00
P.M., and two threads from 5:00 P.M. to midnight. This gives you the ability to fine-tune workloads on your system.
The diagram below shows several different Queues.
Applications Manager 9.4.1
Queue 1 has zero threads from midnight to 8:00 A.M., three threads from 8:00 A.M. to 4:00 P.M., and one thread
from 4:00 P.M. to 23:59:59. Queue 2 has one thread from midnight to 8:00 A.M., and three threads for the rest of
the day.
When Threads Are Set to Zero
If the thread setting for a Queue is zero, any tasks scheduled to run through that Queue will not be launched. The
tasks will be displayed in the Backlog with a status of QUEUE WAIT.
For example, you can set threads to zero to prevent end-user submissions from overwhelming the system during
the working day. To do this, set up a Queue with a schedule that includes zero threads from 7 A.M. to 6 P.M., and
10 threads at all other times. End-users would be able to submit tasks into the Queue during the normal business
day, but the tasks would simply be held until 6 P.M. when the Queue opens up. This way, all tasks submitted during
one day could be run overnight and made available the following day.
Selecting Minimum Threads with Thread Schedules
You can assign a minimum thread value to a Thread Schedule. Any Queue assigned to a Thread Schedule will be
able to have at least that number of threads available in a 'standby' mode.
For example, QUEUE_A (shown below) is guaranteed to have two available threads regardless of any tasks
requested or scheduled in other Queues. If the total threads for the Agent is set to 10, that would mean a maximum
of 8 tasks could run at any time through all the other Queues.
QUEUE_A 1 6 2 2
QUEUE_B 1 7 1 3
QUEUE_C 2 8 3 6
QUEUE_D 2 8 1 7
QUEUE_E 3 10 0 7 3
QUEUE_F 3 10 0 7
Applications Manager 9.4.1
Queues are grouped and run by priority. For example, in the table above you have two Queues in each of the
three priority levels (1, 2, 3). If the Agent Thread limit is set to 10, and the minimum thread limit is set to 2 for
Queue_A and 1 for Queue_B, that leaves only seven threads available to Queues with other priorities. Based on
the Queues used in the example table above, if Queue_F were the only Queue running tasks, only three tasks
would be allowed to run because seven min threads with a higher priority level are already reserved.
Defining Queues
You control the flow of tasks to servers by using Applications Manager Queues. All tasks pass through an
Applications Manager Queue to get to a server. You control Queue throughput by assigning each Queue to a
Thread Schedule.
Applications Manager User Groups control access to Queues. If you do not have access to them, see your
Applications Manager administrator.
Procedure
To add a Queue:
1. From the Queues Selector window, click New.
Applications Manager opens the Queues window shown above.
2. Complete the fields using the information in the following table.
• Name
The name may be up to 30 characters long.
• Thread Schedule
The Thread Schedule controls the maximum number of concurrent tasks that can run through the Queue and
the number of threads that are reserved for the Queue at any given time. For more information, see Defining
Thread Schedules.
• Priority
Defines the order in which each Queue is scanned for task execution if the maximum number of threads for
an Agent is reached. Queues are scanned starting with the number 1 and ending with 99.
The execution order of tasks waiting to run in a QUEUED status is decided in the following order:
1.1 Queue priority
2.1 Job priority
3.1 Start date and time
Therefore, if two tasks are waiting to run in different Queues, and those Queues have the same priority, the
Jobs' priority is checked. If Queue and Job priorities are the same, their start date and times are compared.
• Active
Click the Active checkbox to activate the Queue.
Applications Manager 9.4.1
Active Queues apply the thread settings. When you inactivate a Queue, no tasks will be processed through
it. The tasks will remain in the Backlog with a QUEUE WAIT status until the Queue is activated.
Editing Queues in Explorer
Applications Manager operators with the necessary User Group access can alter Queue definitions from Explorer.
For more information, see Changing Queue Settings.
Applications Manager User Groups control access to Thread Schedules. If you do not have access to them, see
your Applications Manager administrator.
Procedure
To define a Thread Schedule:
1. From the Thread Schedules Selector window, click New.
Applications Manager opens the Thread Schedules window shown above.
2. Enter a name and description for the Thread Schedule on the General tab of the Thread Schedules window
shown in the image above.
Note that there is one entry in the table on the General tab.
Applications Manager 9.4.1
3. If you wish to specify times to set specific minimum and maximum threads, click New.
Applications Manager opens the Thread Schedule details window where you can specify the details for each
entry in your schedule.
4. Complete the fields on the Thread Schedule details window using the information in table below and click OK.
• Start Time
The start time for this entry (hh:mm).
• Stop Time
The stop time for this entry (hh:mm).
• Min Thread
The number of the Automation Engine's threads you wish to reserve for a Queue. The values can range
between 0 and the maximum number of threads available on the Automation Engine. If one or more
minimum threads are assigned to a Queue's Thread Schedule, then even higher priority Queues (Queues
with a lower number set in their Priority field) will have this many fewer threads available to them. This is
even true when no tasks are running in the Queue that is reserving the threads.
• Max Thread
The maximum number of tasks that can run concurrently in the Queue. The value can be between 0 and
9999. If you set the value to 0, the Queue will not accept any tasks during the time period specified. Tasks
assigned to this Queue will show a QUEUE WAIT status.
Only tasks in a RUNNING status count toward the Max thread setting.
Applications Manager adds a new entry to the table with the values you specified. You can add as many entries
to the table as needed to set different minimum and maximum values for different times of the day.
Updating and Deleting Thread Schedule Entries
To update or delete a Thread Schedule entry, select the entry in the table and select Edit or Delete.
Selecting Thread Schedules for Agents
You can select a Thread Schedule for an Agent to specify the maximum number of concurrent tasks that can run on
that Agent at one time in all Queues.
Using the Forecast feature, you can view a list of scheduled Jobs and Process Flows. An example forecast is
shown below.
You can view a graphical forecast of scheduled Jobs and Process Flows. Graphical forecasts are displayed in
the Forecasted Gantt view window shown in the image below. For more information on graphical forecasts, see
Viewing Graphical Forecasts.
The data displayed in forecasts is generated and loaded into Applications Manager by running the FORECAST
Job. When you create a schedule for the FORECAST Job, you determine the time frame of the forecast and how
often it is run. For more information on running the FORECAST Job, see Setting the FORECAST Job Parameters.
You can get a more detailed look at tasks that are scheduled to run by generating a production schedule. An
example production schedule is shown below.
Viewing Forecasts
The Forecast window shows you a list of scheduled Jobs and Process Flows.
Procedure
To open the Forecast window shown above, do one of the following:
• Open the Activities menu and select Forecast.
• Select the Forecast icon from the toolbar.
Applications Manager displays a list of forecasted Jobs and Process Flows shown below.
Each scheduled Job/Process Flow includes the start date and time and the Job or Process Flow's name. Process
Flows also include a key icon used to expand or collapse them.
To view the Jobs within a Process Flow, click the Process Flow's key. To expand all keys for a Process Flow and its
children, select the Process Flow, go to the View menu and select Expand Process Flows. To expand the keys in
all Process Flows, choose Expand All. To print or preview a forecast, use the print buttons.
Applications Manager 9.4.1
You can also select Predecessor Statement to only view the predecessor statement(s) for a Job, Process Flow, or
Process Flow component as shown below.
Data displayed in forecasts is generated and loaded into Applications Manager by running the FORECAST Job.
When you create a schedule for the FORECAST Job, you determine the time frame of the forecast and how often it
is run. For more information on running the FORECAST Job, see Setting the FORECAST Job Parameters.
Procedure
To open the Forecasted Gantt view window shown above.
1. Do one of the following:
• Open the Activities menu and select Graphical Forecast.
• Select the Graphical Forecast icon from the toolbar.
Applications Manager displays the Forecast filter window shown below.
2. If you wish to filter the tasks displayed in the Forecasted Gantt view window, edit the options described below.
Applications Manager 9.4.1
Include minute and hour Not include tasks that are scheduled by minutes or
hours (when unchecked).
Data displayed in the Forecasted Gantt view window is generated and loaded into Applications Manager by
running the FORECAST Job. When you create a schedule for the FORECAST Job, you determine the time
frame of the forecast and how often it is run. For more information on running the FORECAST Job, see Setting
the FORECAST Job Parameters.
By entering information into the fields on the Forecast filter window, you are filtering beginning with the next
scheduled task and ending with the last scheduled task loaded by running the FORECAST Job. You cannot
enter a date/time beyond the setting from the last run of the FORECAST Job.
The Forecasted Gantt view window will not include any tasks that are currently in the Backlog, even if the
tasks are staged. For more information on staging tasks, see chapter Staging Tasks in the Backlog.
3. Click OK.
Applications Manager displays a small animated window as the tasks are loaded for the Forecasted Gantt view
window.
Applications Manager displays the graphical forecast in the Forecasted Gantt view window. For information on
reading Gantt windows, see Reading the Gantt View Window.
4. If you wish, you can right-click a task and select Graphical Predecessors to view its predecessor links in a flow
diagram.
Gantt Legend
The legend describes the graphics used in the Gantt view windows. To display the legend, click the Legend button
in the menu bar. The Legend includes some symbols not used on the Forecasted Gantt view window. To close the
Gantt Legend window, click the X in the title bar.
Depending on your organization's needs, you can schedule this Job to run every day, or several times each day.
Additionally, if you want to set the forecast on an ad hoc basis, you can submit the Job as shown below.
Applications Manager 9.4.1
The levels of sub Process Flows you wish to show in the forecast.
• Minimum Schedule Units
Select whether you want to limit the forecast to list tasks by day, hour, or minute.
If selecting an interval other than -3, you should review your start and end dates because the shorter interval
settings will produce larger reports.
4. Submit the Job and view the output for the SCHPRINT Job.
Sample Production Schedule
{SALES_REPORTS}REGION_C
Abbreviation Definition
The Backlog Gantt view window displays an expandable task tree on the left, and the Gantt chart on the right. You
can change the size of the two panes by dragging the vertical splitter bar that divides the panes.
Each task (Job or Process Flow) is displayed on its own row. Rectangles represent the expected run times of the
tasks: black for Process Flows and blue for Jobs. Arrows drawn between the rectangles indicate predecessor links.
Applications Manager 9.4.1
Process Flows
Process Flows are represented by a rectangle with a black border. The rectangle extends from the scheduled start
time to the scheduled completion time based on the average run time for the Process Flow. The average run time is
based on the sum of the average run times of all tasks in the Process Flow.
When a Process Flow is initiated, a green bar is displayed in the rectangle. The green bar indicates the current
run time for the Process Flow. The green bar is displayed until the Process Flow completes or is killed. When a
Process Flow completes, the run time bar turns black.
If a task in a Process Flow aborts, a red X is placed over the Process Flow name in the task tree. Note a Process
Flow never aborts, only the components in a Process Flow.
From an operations standpoint, you can display only the unexpanded Process Flows in the Gantt chart and easily
monitor the system. If a problem arises with a Process Flow, or you want to see more details about individual
components in a Process Flow, you can expand the Process Flow.
Jobs
Jobs are represented by a rectangle with a blue border. The rectangle extends from the scheduled start time to the
completion time based on the average run time for the Job.
Actual run times for Process Flows and Jobs are represented by solid bars running through the center of the
rectangles. The color of the bar indicates the status of the task.
To help you interpret the start and end time symbols as they relate to Process Flows, several examples are given
below.
Running on schedule
Job Examples
To help you interpret the start and end time symbols as they relate to Jobs, several examples are given below.
Running on schedule
Procedure
To find a task in the Backlog Gantt view window:
1. Go to the Actions menu and select Find.
Applications Manager opens the Find component window shown above. The Find component window lists
all Jobs and Process Flows alphabetically. Process Flow components including sub Process Flows are listed
starting with their initial parent Process Flow.
You can enter a valid UNIX regular expression in the Search field to have Applications Manager filter the list of
tasks. In the image above, .*DAT is entered in the Search field to limit the list of tasks to ones with the letters
'DAT' in their name.
2. Select a task from the Find component window.
Applications Manager highlights the task in the Backlog Gantt view window.
Print Preview
The Print Preview command displays the dialog shown below. You can print directly from this dialog if you wish.
Applications Manager 9.4.1
Print Gantt
The Print Gantt command displays the standard Windows print window shown below where you can set various
output options.
Applications Manager 9.4.1
Status Description/Action
Status Description/Action
Status Description/Action
Status Description/Action
Status Description/Action
BAD TYPE Description: The data type specified for the task is
incorrect.
Action: Check the data type definition assigned to the
prompt.
Status Description/Action
DATE PENDING Description: This task is not yet ready to run. Tasks are
not ready to run if:
• They have been staged by running either the
STAGING or STAGING_BY_SCHEDULE Jobs and
selecting prompts.
• Someone post-dated the start date/time for a Job or
Process Flow when requesting it.
• A condition with a DELAY TASK action prohibits the
task from running.
Action: No action is required. The task will run when its
reaches its start date and time, or when its conditions
allow it to. You can edit the task details of the Job or
Process Flow in the Backlog.
Status Description/Action
HOLD PRED WT Description: The task is on hold and is waiting for one
or more predecessor requirements to be met.
Action: The task will not be eligible to run until both of
the following occur:
• The hold status is changed by an action associated
with another condition or by a User.
• The predecessor requirements for the task are met.
Status Description/Action
Status Description/Action
LAUNCHED Description: The task has been launched, but its status
has not yet been determined.
Action: No action required.
Min Run Time Description: The task finished more quickly than the
time specified in the Min run time field of the Job's
definition. Applications Manager developers specify a
min run time when they expect a Job to run for at least
the time they specify.
Action: Investigate why the task finished so quickly.
Applications Manager 9.4.1
Status Description/Action
Status Description/Action
SELF WAIT Description: The Single run option is set for the Job
or Process Flow, and there is another instance of the
task running. The first must complete before the second
instance will be initiated.
Action: No action required. If you want to allow two
instances of the task to run concurrently (even within
a Process Flow), uncheck the Single run option in the
Job/Process Flow definition.
Status Description/Action
Status Description/Action
7 Development Guide
The Development Guide is a comprehensive procedures manual that covers all aspects of
Applications Manager development.
The Development Guide is a comprehensive procedures manual that covers all aspects of Applications Manager
development. It is part of the complete Applications Manager documentation set which includes:
• Getting Started Guide
• Installation Guide
• User Guide
• Development Guide
• Administration Guide
• Oracle Applications Extension Guide
• PeopleSoft Extension Guide
Text Conventions
The following text conventions are used throughout this guide:
• User interface field names, menu items, and window names are written in bold.
• File names and text within scripts are written in bold.
• Variable text is written within <brackets>. In the example below <run ID number> represents the actual run ID
number of a requested Job.
If you submit a large Process Flow, the message will read, 'Task submission in progress: Run ID = <run ID
number>' until all components of the Process Flow have been placed into the Backlog.
You define Jobs once, but can assign them to multiple Process Flows.
Step 2: Define Objects
As you define Jobs, you use a variety of simple and compound objects. A compound object is made up of one
or more simple objects. You can define objects as needed. Examples of simple and compound objects are listed
below.
Applications Jobs
6. Add scheduling information to the Process Flow by choosing the time and days you want the Process Flow to
run.
If appropriate, add scheduling eligibility to each component in the Process Flow.
7. Save the Process Flow.
After saving the Process Flow, Applications Manager will begin running it on its defined schedule(s).
Only the objects assigned to you via User Groups will be displayed in the selector window. In the image above,
the Queues Selector window is open.
The Jobs and Process Flows selector windows include an Application box. Applications specify a group of
Jobs and Process Flows. The Application you select determines which Jobs and Process Flows are listed in the
table. For more information on Applications, see Defining Applications.
Selector windows are not displayed on the taskbar because they do not contain unique information. For more
information on using the Applications Manager desktop, see the User Guide.
2. To update an existing object, select the object and click Edit. To refresh the list of objects on the selector
window, click Refresh.
You can type the first few letters of an object's name in the Search field, and Applications Manager will find it.
The Search field accepts valid UNIX regular expressions. For example, to search for all Jobs starting with the
letters A and T, you would enter [at] in the Search field. For more information on syntax accepted by regular
expressions, see Appendix A: Regular Expression Tables.
3. To define a new object, click New.
Applications Manager opens an object definition window. If you are defining a new object, the fields will be
empty except where defaults come preselected.
Required fields are marked with a '*' symbol to their right.
4. Enter values for the fields on the various tabs of the object.
When you enter or change a field value, Applications Manager displays a blue triangle next to the current tab's
label to signify unsaved changes. A red triangle signifies unsaved changes made to a sub-element of a tab. For
more information, see Updating Unsaved Changes.
If a supporting object is selected in another object's field, and a User does not have access to the supporting
object, that field will be grayed out. For example, assume the MONITOR Thread Schedule is assigned to the
BATCH Queue. A User named Pat has the BATCH Queue in an edit User Group, but she doesn't have the
Applications Manager 9.4.1
MONITOR Thread Schedule in any User Groups. Pat can edit the BATCH Queue, but she cannot change its
Thread Schedule, because the field is grayed out for her (showing MONITOR as its selection).
5. If you wish, you can select the User Groups tab to assign the object to one or more User Groups. For more
information on assigning objects to User Groups, see the Administration Guide.
If you are assigned to a Maintenance user group, the object will automatically be assigned to your User
Group. You can view a list of objects that you have access to from the View menu by selecting View Assigned
Objects. If you do not have access to the object, see your Applications Manager administrator.
6. Select the appropriate button:
• To save the object's definition and close the window, click OK.
• To add/update the object's definition and keep the window open, click Apply.
• To close the window without updating the object's definition, click Cancel.
Deleting Objects
To delete an object, highlight the object in its selector window and click Delete. If the object you are deleting is
referenced by one or more objects, you must remove the references before it can be deleted. If you try to delete an
object without first removing the references, Applications Manager will display a message saying it is in use. For
information on viewing references for an object, see Viewing Object References.
If Jobs or Process Flows are in the Backlog, you cannot delete their definitions. You will need to wait until they
complete executing.
Prerequisites
Applications Manager 9.4.1
There are several prerequisites to keep in mind when copying Applications Manager objects:
• You must have edit access to the original object. This access is controlled by User Groups. If you created the
original object, you probably have edit access to it.
• You must have User Group access to all the User Groups that are assigned to the object.
• The object you want to copy must be displayed on its object type selector window.
If you are copying an object recently created by another User, it may not be displayed in the selector window. You
may need to refresh the display by clicking the Refresh button.
Copying Objects
To copy an Applications Manager object, select the object in its selector window and choose the Copy button.
For information on using selector windows, see Adding, Editing, and Deleting Applications Manager Objects.
The EXPRESS Queue is selected in the Queues Selector window above. After you choose the Copy button,
Applications Manager opens the new object definition. To save the new object, give it a name (and description, if
required).
Since they are only a name and description, you cannot copy Output Groups.
Copying Jobs and Process Flows
When you copy a Job or Process Flow, the prompts are always copied. You have the option of coping its
conditions, documentation, and schedules.
3. Enter the name and description for the new Job or Process Flow.
4. Select the appropriate checkboxes to copy conditions, documentation, or schedules.
5. To accept the information and close the window, click OK.
Applications Manager closes the Copy Job or Copy Process Flow window, adds the Job or Process Flow to
the selector window, and opens the new Job or Process Flow's window.
You can view Reports for each of the operations windows and selector windows.
Applications Manager 9.4.1
An object type Open that object's selector window and click Reports.
An operations window Open the operations window and choose the Report
type from the Reports menu.
This opens the Reports window and selects the Report type that corresponds to the window you opened it from. In
the image below, the Reports button is selected on the Jobs Selector window, opening the Reports window with
the Jobs type highlighted.
To view Reports for another object type, select that object type from the Type box. If an object is not listed in the
Type box, there are no Reports for it. To view a Report, select it and click the Show button.
Prompt Values
Some Reports require you to enter prompt values. If the Report you select requires prompt values, you must
respond to them in the Report Parameters window shown below. After entering the prompts, click OK and
Applications Manager will display the Report in its own window.
To view or edit the definition a referenced object, select the object and click Edit.
In the Jobs window shown above, unsaved changes have been made on the General tab.
If you click Cancel, Applications Manager displays the window shown below. From this window you can click Yes to
discard the pending changes for the tabs listed, or No to remain on the tab in the object's definition.
Creating Jobs
A Job is the basic building block in Applications Manager. For each program you want to run (such as FTP,
application, or database load), you must create a Job. A Job specifies all the information required to run a program
including:
• General information: the task that will be run and its program information
• Execution options and output settings
• Prompts: information that is passed to the program as variables
Applications Manager 9.4.1
Jobs can be run individually with a schedule or on an ad hoc basis from the Requests window. They also can
be run as a component of a Applications Manager Process Flow. The Jobs window is shown below with the
PROCESS_FLOW_REPORT Job displayed.
Keeping with the Applications Manager object-oriented approach to operations, you can use a Job in as many
different Process Flows as you wish. If you change a Job definition, the change is applied to every Process Flow
that includes the Job. You do not have to change the definition in each Process Flow. This saves you a great deal
of time maintaining your system.
Related Chapters
This chapter focuses on the basics of creating a Job. For additional information, see the following:
• Working with Predecessors
• Adding Prompts to Jobs and Process Flows
• Working with Substitution Variables and Replacement Values
• Working with Conditions
• Scheduling Jobs and Process Flows
Defining Jobs
To add a Job to Applications Manager, go to the Object Admin menu on the desktop and select Jobs.
Applications Manager 9.4.1
Applications Manager User Groups control access to Jobs. If you do not have access to them, see your
Applications Manager administrator.
Jobs and Process Flows should not be edited while running the STAGING Job or executing exports and imports.
For Rapid Automation Agent documentation including Agent-specific Login object tabs, see the the following.
• The Banner Rapid Automation Agent for Local Clients Documentation
• The Banner Rapid Automation Agent for the Automic Web Interface Documentation
• All other Rapid Automation Documentation
Procedure
To add a Job to Applications Manager:
1. From the desktop, open the Jobs Selector window in one of two ways:
• Select the Jobs icon from the toolbar.
• Go to the Object Admin menu and select Jobs.
For more information, see Adding, Editing, and Deleting Applications Manager Objects.
2. From the Jobs Selector window, click New.
Depending on your configuration, you may need to select a Job type from the Select job type window shown
above. The Job type you select will determine the default selections for certain fields, such as the Program Type
and Login.
Applications Manager opens the Jobs window.
To create OAE or PeopleSoft Jobs, go to the Agents Selector window, select the appropriate Agent and use
the Aux button.
3. Define the Job by assigning values to the fields on the General tab. For more information, see Entering General
Information for Jobs.
4. If you wish, you can edit the information on the other tabs in the Jobs window at this time. For information on
each of the tabs of the Jobs window, see Using the Tabs of the Jobs Window.
Applications Manager 9.4.1
5. Click Apply to add the Job definition to the list of Jobs, or OK to add the new definition to the list of Jobs and
close the Jobs window.
6. Applications Manager displays an error message to alert you to any required fields that have not been filled in.
Editing and Deleting Jobs
To edit or delete a Job, select the Job on the Jobs Selector window and select the appropriate button. For
information on using selector windows, see Adding, Editing, and Deleting Applications Manager Objects.
If a Job is used in one or more Process Flows, you must remove it before you can delete the Job. If a Job is used in
a Process Flow and you try to delete it, Applications Manager will display a message saying the Job is in use.
If a task for a Job is in the Backlog, you cannot delete it. You will need to wait until the task has completed
executing before you can delete its definition.
Some of the tabs are more commonly used than others. Additionally, the Process Flows window includes several
of the same tabs as the Jobs window. For these reasons, some tabs on the Jobs window are described in this
chapter and others are described in later chapters.
Job Tabs Discussed in this Chapter
The following tabs are discussed in this chapter:
• General: Specifies the name and description of the Job. It also tells Applications Manager where to find the
program or script the Job will run and how to run it. Many of the fields on this tab are required. For details, see
Entering General Information for Jobs.
• Execution Options: Determine what happens when a Job runs. For details, see Entering Execution Options for
Jobs.
• Output: Controls what happens to the output of the Job. For details, see Specifying Output Options for Jobs.
• Documentation: Provides a location to enter relevant information about the processing of a task. By selecting
an individual Job and choosing Documentation, operators can access these comments, suggestions, or
instructions for an individual task. For details, see Adding Job Documentation.
• User Groups: Controls access to the Job. User Groups can be added to a Job from the User Groups tab as
well as from the User Group itself. User Groups are usually defined by your Applications Manager Administrator.
For details, see Assigning User Groups to Jobs.
Job Tabs Discussed in Other Chapters
The following tabs are discussed in later chapters:
• Schedule: Creates schedules to run the Job. Schedules can be based on days of the week, specific days of
the month, and days in a Calendar. You create more than one schedule for a Job. For details, see chapter
Scheduling Jobs and Process Flows.
• Prompts: Pass arguments or parameters to a program or script. For more information, see chapter Adding
Prompts to Jobs and Process Flows.
• Predecessors: Specify external predecessor links for the Job. External predecessor links are dependencies
on other stand-alone Jobs, Process Flows, or Process Flow components. While predecessors are used in all
Process Flows, they are only defined for Jobs in special case scenarios. For details, see chapter Working with
Predecessors.
• Conditions: Conditions control the execution of tasks. They provide IF/THEN logic for your Job. Conditions
can be evaluated before, during, and after a task executes, or after a task is deleted. For details, see chapter
Working with Conditions.
Applications Manager 9.4.1
>cat employees.sql
set verify off
set feedback off
set termout off
spool &so_outfile
column ename heading 'Employee|Name'
column dname heading 'Dept|Name'
select emp.ename, dept.dname
from emp, dept
where dept.dname = '&dept_name'
and emp.deptno = dept.deptno;
spool off
When the variable dept_name is passed to the employees.sql script, it returns all the employees in that
department.
Creating the Job
The first step in creating a Job to run the employees.sql script is to define a new Job and fill in the information on
the Job's General tab. A sample Job named EMPLOYEES that runs the employees.sql script is shown below.
Applications Manager 9.4.1
The specifics of all the fields on the General tab are discussed in later topics. For now, note that there is a
Program box that tells Applications Manager what to run. The path in the Location field is based on selections
made in each of the fields in this box Note that AWSQLP is selected in the Type field. This designates the Job as a
SQL*Plus Job.
Assigning a Login to the Job
Because SQL tasks need to access a database, a Database Login must be assigned to the Job on the Execution
Options tab as shown below.
Applications Manager 9.4.1
Applications Manager Logins give users access to a database or Host Login without requiring that they know the
password.
Adding a Prompt to the Job
To allow users to pass a dept_name variable to the employees.sql script, you need to add a prompt to the Job as
shown below.
Applications Manager 9.4.1
The prompt above uses the Dept_Name data type to pass a value to the dept_name variable. The Dept_Name
data type includes a SQL statement that allows users to select a single entry from a list of departments.
Good to Go
With the required fields completed on the General, Login, and Execution Options tabs, the Job is ready to run
from the Requests window. It can also be added to a Process Flow object.
You can define certain objects by clicking the icon to the right of their field. Applications Manager will display the
selector window where you can define the new object. For example, you can click the Application icon to display
the Application selector window.
Entering Job Header Information
All values in the Job header are required. They are described below:
• Name
The name may be up to 30 characters long. Job names can include letters, numbers, periods, colons,
underscores, and hyphens. The first letter of a Job name must be a letter or number.
Do not name Jobs 'SCHED-<Process Flow name>', unless they are Process Flow requestor Jobs.
• Description
The description may be up to 100 characters long.
• Application
Applications are the general category to which Jobs and Process Flows belong (for example: inventory,
accounts payable). They are used on the Requests window to limit the list of Jobs and Process Flows. For
more information on defining Applications, see Defining Applications.
• Agent/Group
The Agent where the program will execute. This field allows you to select Agents and Agent Groups. To run the
Job on:
• A single machine defined as an Agent, select the Agent.
• Either of two or more Agents assigned to an Agent Group, select the regular Agent Group.
Applications Manager 9.4.1
• All Agents assigned to an Agent Group, select the multi-execution Agent Group.
For information on which Agents are assigned to what groups, or which Agent Groups are multi-execution, see
your Applications Manager administrator.
When Agent Groups are assigned to Jobs, particular Agents may be specified for Process Flows, Process Flow
components, requests, or in schedules. For more information, see How Agent Assignments Are Handled for
Process Flow Components.
• Queue
The Job will run on this Queue even when added to the Process Flow, unless the Insert components into
process flow's queue Automation Engine option is set. For more information on setting Automation Engine
options, see the Administration Guide.
If the Job is run using a schedule and that schedule includes a Queue, that setting overrides the Job's Queue.
If the Job is submitted on an ad hoc basis, the User may have the option to change the Queue on the Submit
window.
• Single run
When selected, two instances of the Job cannot run concurrently. The second instance will show a status of
SELF WAIT until the first completes..
• Active
When selected, the Job will run when scheduled or included in another Process Flow and be available from the
Request window. If not selected, the Job will not run when scheduled or included in another Process Flow, and
it cannot be run from the Requests window.
• Stay in queue on abort
When selected, the aborted Job remains in the Backlog so it can be examined, repaired, and reset. The Queue
is not blocked by this aborted Job, unless the Queue has only one thread and the Allow aborted task to block
single threaded queue Automation Engine option is checked.
When this option is not checked, and the Job aborts and leaves the Backlog, it will satisfy any Success
predecessors it may have with other tasks. Failure and Failure (skip on success) predecessors will not be
satisfied.
If this option is not selected and the Job aborts, it leaves the Backlog and a record is written to History. The
Job may be examined from History to determine why it aborted, but because there is no longer a record in the
Backlog, it cannot be repaired or reset.
When this option is selected, and the Restart once on abort option is not selected, and the Job aborts, you will
see the original Run ID <run_id> in History and the current listing <run_id>.01 in the Backlog.
This setting can be overridden with a condition. This is most commonly done for Process Flow components.
Entering Program Information
In order for a Job to execute a program or run a Process Flow, you must define its program information in the
Program box on the General tab. Jobs created for some Applications Manager extensions such as OAE and
PeopleSoft may have different fields in their Program box than what is shown in the image above for standard
Jobs. For more information, see your Applications Manager extension documentation.
The Program fields for standard Jobs are described below:
• Library
The path to the program source or executable. The Program Type defined in the Type field may supply a
subpath. For information on defining Libraries, see Defining Libraries.
• Library path
This non edit field displays the path of the Library selected in the Library field.
• Program Type
This defines which type of program the Job will run, such as shell scripts, host command, or SQL*Plus.
Applications Manager comes with several Program Types already defined. For more information on defining
Program Types, see Defining Program Types.
• Program name
The name of the program that will be executed by the Job. Program names can be typed into this field, or
selected based on the Job's Agent, Library, and Program Type using the Select button. When you click the
Select button, Applications Manager displays a list of valid file names in the path defined by the Agent, Library,
Applications Manager 9.4.1
and Program Type. If you use the Select button for a Job assigned to an Agent Group, you will be able to
search for the file on the Agent you select. If you type in a program name, Applications Manager does not verify
that it exists.
• Location
As you enter the program information, Applications Manager displays the path and file name in this non-editable
field.
Defining Applications
The Application field provides a way to assign Jobs as well as Process Flows to an organizational group. The
groups might be programs, machines, divisions, or some other characteristic. When editing or requesting Jobs and
Process Flows, you can select an Application and the Jobs and Process Flows displayed will be limited to those
that are assigned to the Application.
Applications Manager User Groups control access to Applications. If you do not have access to them, see your
Applications Manager administrator.
Procedure
To define a new Application:
1. From the Applications Selector window, click New.
Applications Manager opens the Applications window shown above.
For information on using selector windows, see Adding, Editing, and Deleting Applications Manager Objects.
2. Complete the fields in the Applications window using the information below.
• Name
The name may be up to 30 characters long.
• Description
The description may be up to 100 characters long.
• Notification
Optionally use Notifications to send messages and output files based on task status to email addresses or
any Output Device defined in Applications Manager. For information on defining Notifications, see Defining
Notifications.
• Environment Variables
Optionally specify one or more Environment Variables as a single Applications Manager object. For
information on defining Environment Variables, see Defining Environment Variables.
Applications Manager 9.4.1
Defining Libraries
When you create a Job, you specify the location of the program or script to run on a host. The Library defines a
path, which will be the first part of that location. A subdirectory of the Library path can be specified when you select
a Program Type for the Job. You can use the same Library for Jobs that run on different operating systems by
specifying a path for each.
Applications Manager User Groups control access to Libraries. If you do not have access to them, see your
Applications Manager administrator.
Procedure
To define a new Library:
1. From the Libraries Selector window, click New.
Applications Manager opens the Libraries window shown above.
For information on using selector windows, see Adding, Editing, and Deleting Applications Manager Objects.
2. Enter a Library name.
The name is a short identifier for the path.
3. Enter a path for one or more operating systems.
Library paths can be hard-coded or they can include Environment Variables. The advantages of using
Environment Variables are:
• The same variable can be defined with different paths on all Agents of a particular operating system.
• You can export and import Libraries and be assured that the Jobs using them will point to the right path in the
other Applications Manager instance.
When Environment Variables are used, you must define the variable for each Applications Manager Agent.
You cannot use Environment Variables that are defined with Applications Manager Environment Variable
objects.
For UNIX machines, define the variable at the end of the $AW_HOME/site/sosite file in Bourne Shell format. To
define a variable called INVERTORY_APP, you would enter:
For Windows machines, define the variable at the end of site/envvar.bat as follows:
set INVENTORY_APP=\inventory\program
If you wish to modify one of the Libraries that ships with Applications Manager, create a new Library using the
Copy function. The Libraries that ship with Applications Manager will be overwritten each time you upgrade.
Applications Manager User Groups control access to Program Types. If you do not have access to them, see your
Applications Manager administrator.
Multiple Operating Systems Supported
Each Program Type can support UNIX and Windows operating systems if the corresponding fields in the OS
portion of the General tab are completed. This eliminates the need to define separate Program Types for each
OS as required in previous versions of Applications Manager. In the example shown above, the same Program
Type script (host command) is used for all three operating systems, but the syntax used in the command paths are
different. Other interfaces might require different scripts for each OS, and there