data services management console administrator guide3291

Data Services Management Console:Administrator Guide

BusinessObjects Data Services XI 3.0 (12.0.0)

© 2008 Business Objects. All rights reserved. Business Objects owns the followingU.S. patents, which may cover products that are offered and licensed by Business

Copyright

Objects: 5,555,403; 5,857,205; 6,289,352; 6,247,008; 6,490,593; 6,578,027;6,831,668; 6,768,986; 6,772,409; 6,882,998; 7,139,766; 7,299,419; 7,194,465;7,222,130; 7,181,440 and 7,181,435. Business Objects and the Business Objectslogo, BusinessObjects, Business Objects Crystal Vision, Business Process OnDemand, BusinessQuery, Crystal Analysis, Crystal Applications, Crystal Decisions,Crystal Enterprise, Crystal Insider, Crystal Reports, Desktop Intelligence, Inxight,the Inxight Logo, LinguistX, Star Tree, Table Lens, ThingFinder, Timewall, Letthere be light, Metify, NSite, Rapid Marts, RapidMarts, the Spectrum Design, WebIntelligence, Workmail and Xcelsius are trademarks or registered trademarks inthe United States and/or other countries of Business Objects and/or affiliatedcompanies. All other names mentioned herein may be trademarks of their respectiveowners.

Business Objects products in this release may contain redistributions of softwarelicensed from third-party contributors. Some of these individual components may

Third-partyContributors

also be available under alternative licenses. A partial listing of third-partycontributors that have requested or permitted acknowledgments, as well as requirednotices, can be found at: http://www.businessobjects.com/thirdparty

2008-03-14

ContentsIntroduction 9Chapter 1

Welcome to Data Services........................................................................10Welcome..............................................................................................10Documentation set for Data Services...................................................10Accessing documentation....................................................................13Business Objects information resources..............................................14

Overview of this guide...............................................................................16About this guide....................................................................................16Who should read this guide..................................................................17

Administrator User Interface 19Chapter 2

Installation and configuration.....................................................................20About the Management Console...............................................................22

Logging in.............................................................................................23Management Console navigation.........................................................24

Administrator navigation............................................................................25Navigation tree.....................................................................................25Pages...................................................................................................29

Administrator Management 31Chapter 3

Adding repositories....................................................................................32Connecting repositories to the Administrator.......................................33Changing repository connection allocation..........................................36Adapter considerations.........................................................................36

Managing user roles..................................................................................37To add users and their roles.................................................................38

Data Services Management Console: Administrator Guide 3

Adding Access Servers..............................................................................39To connect an Access Server to the Administrator..............................40

Centralizing administration........................................................................40To group administration by job type.....................................................41

Setting the status interval..........................................................................41To set the status interval.......................................................................41

Setting the log retention period..................................................................42To delete log information......................................................................42

Managing database account changes.......................................................43Updating local repository login parameters..........................................44Updating datastore connection parameters.........................................45

Central Repository Management 47Chapter 4

Setting up users and groups......................................................................48To add the secure central repository to the Administrator....................48To add a group to a central repository..................................................49To add users.........................................................................................49To add or remove a user from a group.................................................51

Deleting groups..........................................................................................51To delete a group..................................................................................52

Viewing reports..........................................................................................52Object state report................................................................................52Change report......................................................................................53

Server groups 55Chapter 5

Server group architecture..........................................................................56Load balance index..............................................................................58Job execution.......................................................................................58Job launcher.........................................................................................59Working with server groups and Designer options...............................59

4 Data Services Management Console: Administrator Guide

Contents

Editing and removing a server group.........................................................61To edit a server group...........................................................................62To remove a server group....................................................................62

Monitoring Job Server status in a server group.........................................63Executing jobs using server groups...........................................................64

Batch Jobs 65Chapter 6

Executing batch jobs..................................................................................66To execute a job...................................................................................66

Scheduling jobs.........................................................................................67Using the Data Services job scheduler................................................67Scheduling jobs in BusinessObjects Enterprise...................................79Using a third-party scheduler...............................................................82Data Services job launcher..................................................................88

Monitoring jobs..........................................................................................91Overall status.......................................................................................91Statistics...............................................................................................93Ignore error status................................................................................96Deleting batch job history data.............................................................96Stopping a running job.........................................................................97Trace, monitor, and error logs..............................................................97

Real-Time Jobs 99Chapter 7

Supporting real-time jobs.........................................................................100Configuring and monitoring real-time services........................................103

Creating services and service providers............................................103Starting and stopping services...........................................................109Updating service providers.................................................................112Monitoring services............................................................................113

Creating and monitoring client interfaces................................................115


Contents

RFC clients.........................................................................................116Adding IDoc configurations to an RFC client.....................................119Message Broker clients......................................................................121Monitoring clients...............................................................................121

Real-Time Performance 125Chapter 8

Configuring Access Server output...........................................................126To configure an Access Server...........................................................127

Service configuration parameters............................................................129Service startup behavior.....................................................................130High traffic behavior...........................................................................131Response time controls......................................................................132

Service statistics......................................................................................133Service provider statistics........................................................................135Using statistics and service parameters..................................................136

Profile Server Management 139Chapter 9

Defining the profiler repository.................................................................140To define a profiler repository.............................................................140

Connecting repositories to the Administrator ..........................................141To add a local, central, or profiler repository connection to theAdministrator......................................................................................141To view the list of repositories connected to the Administrator..........143

Defining profiler users..............................................................................143To define a profiler user......................................................................144

Configuring profiler task parameters.......................................................145To configure profiler task parameters ................................................146Task execution parameters................................................................146Task management parameters...........................................................148

Monitoring profiler tasks using the Administrator ....................................150


Contents

To monitor a profiler task in the Data Services Administrator............151

Adapters 157Chapter 10

Overview of adapters...............................................................................158To create an adapter datastore connection in the Designer...............159

Adding and configuring adapter instances..............................................160To add an adapter instance................................................................160To edit an adapter's configuration......................................................161Adapter instance configuration information........................................161

Starting and stopping adapter instances.................................................165To start an adapter instance...............................................................165To stop an adapter instance...............................................................165To start or stop an adapter operation instance...................................165

Monitoring adapter instances..................................................................166To monitor the adapter instances and operations..............................166To monitor adapter instance statistics................................................167

Support for Web Services 169Chapter 11

Support for HTTP 171Chapter 12

Overview..................................................................................................172Adapter installation and configuration.....................................................173

URL for HTTP requests to Data Services..........................................173Configuring the HTTP adapter...........................................................173Configuring an operation instance.....................................................176Defining the HTTP adapter datastore................................................179Configuring SSL with the HTTP adapter............................................182

Troubleshooting 187Chapter 13

Reestablishing network connections.......................................................188


Contents

To reestablish network connections for your repository.....................188Reinstalling the Web Server service........................................................188

To reinstall the Web Server service....................................................189Finding problems.....................................................................................189

To determine which object is not operating properly..........................190To determine the cause of the error...................................................191

Error and trace logs.................................................................................191Batch job logs.....................................................................................191Service provider logs..........................................................................193Access Server logs.............................................................................195Adapter logs.......................................................................................198

Resolving connectivity problems.............................................................198Restarting the Access Server..................................................................200

To perform a controlled restart of the Access Server.........................201To perform an abort and restart of the Access Server.......................201

Index 203


Contents

Introduction

1

Welcome to Data Services

Welcome

Data Services XI Release 3 provides data integration and data qualityprocesses in one runtime environment, delivering enterprise performanceand scalability.

The data integration processes of Data Services allow organizations to easilyexplore, extract, transform, and deliver any type of data anywhere acrossthe enterprise.

The data quality processes of Data Services allow organizations to easilystandardize, cleanse, and consolidate data anywhere, ensuring that end-usersare always working with information that's readily available, accurate, andtrusted.

Documentation set for Data Services

You should become familiar with all the pieces of documentation that relateto your Data Services product.

What this document providesDocument

Information about available Data Services books,languages, and locationsDocumentation Map

Highlights of key features in this Data Services re-leaseRelease Summary

Important information you need before installing anddeploying this version of Data ServicesRelease Notes

An introduction to Data ServicesGetting Started Guide

Information about and procedures for installing DataServicesInstallation Guide


Introduction1 Welcome to Data Services


Guidelines and options for migrating applications in-cluding information on multi-user functionality andthe use of the central repository for version control

Advanced Development Guide

Information about how to use Data Services DesignerDesigner Guide

Information for third-party developers to access DataServices functionalityIntegrator's Guide

Information about how to use Data Services Adminis-trator

Management Console: AdministratorGuide

Information about how to use Data Services MetadataReports

Management Console: Metadata Re-ports Guide

Release-specific product behavior changes fromearlier versions of Data Services to the latest releaseMigration Considerations

Information about how to migrate from Data Qualityto Data ServicesMigration Guide

Information about how to improve the performanceof Data ServicesPerformance Optimization Guide

Detailed reference material for Data Services Design-erReference Guide


1IntroductionWelcome to Data Services


A compiled “master” PDF of core Data Services bookscontaining a searchable master table of contents andindex:• Installation Guide• Getting Started Guide• Designer Guide• Reference Guide• Management Console: Metadata Reports Guide• Management Console: Administrator Guide• Performance Optimization Guide• Advanced Development Guide• Supplement for J.D. Edwards• Supplement for Oracle Applications• Supplement for PeopleSoft• Supplement for Siebel• Supplement for SAP

Technical Manuals

A step-by-step introduction to using Data ServicesTutorial

In addition, you may need to refer to several Adapter Guides andSupplemental Guides.


Information about how to install, configure, and use the DataServices Adapter for JMSJMS Adapter Interface

Information about how to install, configure, and use the DataServices Salesforce.com Adapter Interface

Salesforce.com AdapterInterface

Information about license-controlled interfaces between DataServices and J.D. Edwards World and J.D. Edwards OneWorld

Supplement for J.D. Ed-wards

Information about the license-controlled interface between DataServices and Oracle Applications

Supplement for Oracle Ap-plications




Information about license-controlled interfaces between DataServices and PeopleSoftSupplement for PeopleSoft

Information about license-controlled interfaces between DataServices, SAP ERP and R/3, and SAP BI/BWSupplement for SAP

Information about the license-controlled interface between DataServices and SiebelSupplement for Siebel

Accessing documentation

You can access the complete documentation set for Data Services in severalplaces.

Note: For the latest tips and tricks on Data Services, access our KnowledgeBase on the Customer Support site at http://technicalsupport.businessobjects.com. We have posted valuable tips for getting the most out of your DataServices product.

Accessing documentation on Windows

After you install Data Services, you can access the documentation from theStart menu.

1. Choose Start > Programs > BusinessObjects XI 3.0 >BusinessObjects Data Services > Data Services Documentation.

Note: Only a subset of the documentation is available from the Startmenu. The documentation set for this release is available inLINK_DIR\Doc\Books\en.

2. Click the appropriate shortcut for the document that you want to view.



Accessing documentation on UNIX

After you install Data Services, you can access the online documentation bygoing to the directory where the printable PDF files were installed.

1. Go to LINK_DIR/doc/book/en/.2. Using Adobe Reader, open the PDF file of the document that you want

to view.

Accessing documentation from the Web

You can access the complete documentation set for Data Services from theBusiness Objects Customer Support site.

1. Go to www.businessobjects.com2. From the "Support" pull-down menu, choose Documentation.3. On the "Documentation" screen, choose Product Guides and navigate

to the document that you want to view.

You can view the PDFs online or save them to your computer.

Business Objects information resources

Customer support, consulting, and training

A global network of Business technology experts provides customer support,education, and consulting to ensure maximum business intelligence benefitto your business.

Useful addresses at a glance

ContentAddress

Information about the full range of BusinessObjects products.

Product information

http://www.businessobjects.com





ContentAddress

Business Objects product documentation, in-cluding the Business Objects DocumentationRoadmap.

Product documentation

http://www.businessobjects.com/support

Send us feedback or questions about yourBusiness Objects documentation.

Documentation mailbox

[email protected]

Information on Customer Support programs,as well as links to technical articles, down-loads, and online forums.

Online Customer Support


An online resource for sharing and learningabout Data Services with your developer col-leagues.

Online Developer Community

http://diamond.businessobjects.com/

Information about how Business Objects canhelp maximize your business intelligence in-vestment.

Consulting services

http://www.businessobjects.com/services/consulting/

Information on Business Objects training op-tions and modules.

Education services

http://www.businessobjects.com/services/training

Online Customer Support

The Business Objects Customer Support web site contains information aboutCustomer Support programs and services. It also has links to a wide rangeof technical information including Knowledge Base articles, downloads, andsupport forums. http://www.businessobjects.com/support

Looking for training options?

From traditional classroom learning to targeted e-learning seminars, BusinessObjects can offer a training package to suit your learning needs and preferred














learning style. Find more information on the Business Objects Educationweb site: http://www.businessobjects.com/services/training

Send us your feedback

Do you have a suggestion on how we can improve our documentation? Isthere something that you particularly like or have found useful? Drop us aline, and we will do our best to ensure that your suggestion is considered forthe next release of our documentation: [email protected].

Note: If your issue concerns a Business Objects product and not thedocumentation, please contact our Customer Support experts.

Overview of this guide

About this guide

The guide covers the BusinessObjects™ Data Services Administrator, aweb-based application written entirely in Java. You can install the DataServices Administrator on a separate computer from the other Data Servicescomponents. It runs on the Data Services Web Server, which is supportedby the Data Services Web Server service. The Administrator uses a JDBCconnection to repositories.

Use the Administrator to:

• Set up users and their roles

• Add connections to Access Servers and repositories

• Manage the retention of Job Server and Access Server logs

• Access job data published for Web Services

• Schedule and monitor batch jobs

• Configure and monitor:

• Access Server status

• Real-time services

• Client interfaces including SAP ERP and R/3 client interfaces (to readIDocs) and message traffic moving in and out of an Access Server


Introduction1 Overview of this guide


• Adapter instances (a prerequisite for creating adapter datastores)

Who should read this guide

This and other Data Services product documentation assume the following:

• You are an application developer, consultant or database administratorworking on data extraction, data warehousing, data integration, or dataquality.

• You understand your source and target data systems, DBMS, legacysystems, business intelligence, and messaging concepts.

• You understand your organization's data needs.

• You are familiar with SQL (Structured Query Language).

• If you are interested in using this product to design real-time processingyou are familiar with:

• DTD and XML Schema formats for XML files

• Publishing Web Services (WSDL, HTTP/S and SOAP protocols, etc.)

• You are familiar with Data Services installation environments: MicrosoftWindows or UNIX.


1IntroductionOverview of this guide


Introduction1 Overview of this guide

Administrator User Interface

2

This section describes the Administrator and how to navigate through itsbrowser-based, graphical user interface.

Related Topics• Installation and configuration on page 20• About the Management Console on page 22• Administrator navigation on page 25

Installation and configuration• General information about the components and architecture of Data

Services

• Complete installation instructions for all Data Services componentsincluding connectivity testing for Data Services real-time functionality.

A summary of the connections used in Data Services is included here foryour reference. You must create the connections in the first four rows of thefollowing table before you can log in to the Administrator.

Purpose of this connectionConnection TypeConnection ToolComponent

• Connects Designer andrepositories.

• Provides location forstorage of Data Ser-vices tables and jobmetadata.

• Connection informationis based on thedatabase you use for arepository.

Custom connectionRepository ManagerRepository(s)


Administrator User Interface2 Installation and configuration


Connects a Job Server tothe Data Services Serviceand repository you specify.

You can also set a JobServer to support adaptersvia a separate communica-tion port (default 4001).Required to use theAdapter Instance node inthe Administrator.

Default (3500) orcustom portServer ManagerJob Server(s)

Connects an Access Serv-er to the Data ServicesService and provides a portfor Message Client libraries(allows applications tocommunicate with DataServices).

Default (4000) orcustom portServer ManagerAccess Server(s)

Provides an HTTP port(28080) for connection be-tween Administrator and allAccess Servers.

Includes an automaticallyassigned shutdown port(22828) which is not dis-played. It is used by theTomcat service to start andstop the application server,which supports the Admin-istrator.

For web application serversother than the packagedTomcat server, the portsmay vary.

Automatically as-signed portsInstallerAdministrator


2Administrator User InterfaceInstallation and configuration


Connects Access Server(s) to the Administrator

Computer name onwhich an AccessServer is installedand port (you speci-fied in Server Man-ager). For example:AT589:4000

AdministratorAccess Server(s)

Connects repositories tothe Administrator. JobServers (previously con-nected to each repositoryusing the Server Manager)also link to the Administra-tor with this connection.

Settings based oneach repository'sdatabase

AdministratorRepository(s)

For more information, see the BusinssObjects Data Services Designer Guide.

About the Management ConsoleThe Management Console is a collection of Web-based applications foradministering Data Services jobs and services, viewing object relationships,and evaluating job execution performance and data quality.

These applications include:

• Administrator — Manage your production environment including batchjob execution, real-time services, Web services, adapter instances, servergroups, central and profiler repositories, and more. This guide describesthe Administrator.

• Impact and Lineage Analysis — Analyze the end-to-end impact andlineage for Data Services tables and columns and BusinessObjectsEnterprise objects such as universes, business views, and reports.

• Operational Dashboards — View dashboards of Data Services jobexecution statistics to see at a glance the status and performance of yourjob executions for one or more repositories over a given time period.


Administrator User Interface2 About the Management Console

• Data Quality Dashboards — Evaluate the reliability of your target databased on the validation rules you created in your Data Services batchjobs to quickly review, assess, and identify potential inconsistencies orerrors in source data.

• Auto Documentation — View, analyze, and print graphicalrepresentations of all objects as depicted in Data Services Designerincluding their relationships, properties, and more.

Related Topics• Management Console—Metadata Reports Guide: Impact and LineageAnalysis Reports

• Management Console—Metadata Reports Guide: Operational DashboardReports

• Management Console—Metadata Reports Guide: Data ValidationDashboard Reports

• Management Console—Metadata Reports Guide: Auto DocumentationReports

Logging in

To access the Administrator, first log in to the Data Services ManagementConsole. The first time you log in to the Management Console, use the defaultuser name and password (admin/admin). Business Objects recommendsthat you change the defaults thereafter.

Related Topics• Managing user roles on page 37

To log in to the Management Console

1. The first step depends on your operating system:• On Windows, click the Start menu and select Programs > Business

Objects XI 3.0 > BusinessObjects Data Services > Data ServicesManagement Console.

If you encounter an error, check to see if your web application serveris installed and running.


2Administrator User InterfaceAbout the Management Console

• On UNIX, open a browser, enter the following case-sensitive URL,then press Enter:

http://hostname:28080/DataServices

2. Enter the default user name (admin) and password (admin) and click Login.

The Management Console home page opens.

3. To launch the Administrator, click the Administrator icon (or name).

The Administrator Status page displays a status overview of all jobs.

If you are logged in to the Designer, you can also access the ManagementConsole home page as follows:

• From the Start page, click Management Console.

• From the Tools menu, click Management Console.

• Click the Management Console tool bar icon.

Management Console navigation

After logging in to the Management Console and launching one of theapplications, the application name appears under the Management Consolebanner.

The upper-right side of the main window includes the following links:

• Home—Click to return to the Management Console home page (forexample to select another application).

• Logout—Click to exit the application and the Management Console andreturn to the login page.

• Settings—The metadata reporting applications also include a Settingscontrol panel for changing a variety of options depending on theapplication.

As you navigate around the applications, notice the top of the right-handpane often displays a "bread crumb" path to indicate where you are in theapplication. Depending on the page displayed, sometimes you can click onthe bread crumbs to navigate to a different part of the application.


Administrator User Interface2 About the Management Console

The Administrator, Impact and Lineage Analysis, and Auto Documentationapplications also use a navigation tree in the left-hand pane.

Data Services Management Console sessions time out after 120 minutes (2hours) of inactivity.

Administrator navigationThe layout of the Data Services Administrator consists of a window with anavigation tree on the left and pages with tabs on the right.

Navigation tree

The navigation tree is divided into nine nodes: Status, Batch, Real-Time,Web Services,Adapter Instances, Server Groups,Central Repositories,Profiler Repositories, and Management.

Status node

When the Administrator opens, it displays the Status page. The Status pagedisplays the status of the following items (after you have connected them tothe Administrator). The red, green, and yellow icons indicate the overallstatus of each item based on the jobs, services, and other objects theysupport.

• Batch—Contains the name of the repository associated with Job Serveron which you run the batch jobs. To see batch jobs status, connect therepository to the Administrator.

Click the repository name to display a list of batch jobs and their status.

• Real-time—Contains the name of the Access Servers associated with areal-time service. To see real-time jobs status, connect the Access Serverto the Administrator.

Click the Access Server name to display a list of real-time services andtheir client interfaces.

• Adapters—Contains the name of the repository associated with JobServer on which you run the adapter. To see an adapter's status, enable


2Administrator User InterfaceAdministrator navigation

a Job Server for adapters, then add the repository associated with thatJob Server.

• Profiler—Contains the name of the repository associated with the ProfilerServer. To see a profiler repository, connect the profiling repository to theAdministrator.

Click the repository name to display a list of profiler tasks and their status.

Batch node

After you add at least one repository connection to the Administrator, youcan expand theBatch node. Then click a repository name to display its BatchJob Status page.

Click the All Repositories option to see jobs in all repositories connectedto this Administrator (only appears if more than one repository is connected).

Each repository under the Batch node includes the following tabs:

• Batch Job Status—View the status of the last execution and in-depthinformation about each job

• Batch Job Configuration—Configure execution and scheduling optionsfor individual jobs

• Repository Schedules—View and configure schedules for all jobs in therepository

• Datastore Configurations—Edit some options for a datastore or aparticular datastore configuration rather than using the Designer.

• Resource Management—Manage data transfer and communicationresources that Data Services uses to distribute data flow execution.

Related Topics• Batch Jobs on page 65

Real-Time node

After you add a connection to an Access Server in the Administrator, youcan expand the Real-Time node. Expand an Access Server name under theReal-Time node to view the options.


Administrator User Interface2 Administrator navigation

DescriptionAccess Server nodeoptions

View status of real-time services and client interfacessupported by this Access Server. Control, restart, andset a service provider interval for this Access Server.

Status

View status for services and service providers, startand stop services, add or remove a service, configureJob Servers for a service.

Real-time Services

View status for client interfaces, start and stop inter-faces, add or remove an interface.Client Interfaces

View list of current Access Server logs, content ofeach log, clear logs, configure content of logs fordisplay, enable or disable tracing for each AccessServer.

Logs - Current

View list of historical Access Server logs, view contentof each log, delete logs.Logs - History

Related Topics• Real-Time Jobs on page 99• Real-Time Performance on page 125

Web Services node

Use this node to select real-time and batch jobs that you want to publish asWeb service operations and to monitor the status of those operations. Youcan also use the node to set security for jobs published as Web serviceoperations and view the WSDL file that Data Services generates.

Related Topics• Support for Web Services on page 169



Adapter Instances node

Use this node to configure a connection between Data Services and anexternal application by creating an adapter instance and dependentoperations. This is a prerequisite requirement for creating a datastore foradapters in the Designer.

After you create a datastore, import data through the adapter and createjobs. Then use this node to view the status of Adapter instances. Optionsare listed by Job Server under the Adapter Instance node.

Related Topics• Adapters on page 157

Server Groups node

The Server Groups node allows you to group Job Servers that are associatedwith the same repository into a server group.

Use a server group if you want Data Services to automatically use the JobServer on a computer with the lightest load when a batch job is executed.This functionality improves load balancing (throughput) in productionenvironments and also provides a hot backup method. When a job islaunched, if a Job Server is down, another Job Server in the same groupexecutes the job.

Related Topics• Server groups on page 55

Central Repositories node

The Central Repositories node has configuration options for secure centralrepositories including:

• Users and groups — Use to configure groups and users for secureaccess to central repository objects

• Reports— Use to generate reports about central repository objects suchas which objects a user currently has checked out or the changes madeto an object over a specified time frame



Related Topics• Central Repository Management on page 47

Profiler Repositories node

After you connect a profiler repository to the Administrator, you can expandProfiler Repositories node. Click a repository name to open the ProfilerTasks Status page.

Related Topics• Profile Server Management on page 139

Management node

The Management node contains the configuration options for theAdministrator application. Before you can use the Administrator, you mustadd connections to other Data Services components using the Managementnode. For example, expand the management node and:

• Click Repositories to add a connection to the repositories that containthe jobs and data profiler tables with which you want to work.

• Click Access Servers to add a connection to your Access Servers (forreal-time jobs).

Related Topics• Administrator Management on page 31

Pages

The top of the page indicates the currently selected node. Once you selecta branch on the navigation tree to go to a page, use the tab row on the pageto navigate further.

As you drill into various pages, a "bread crumb" trail often indicates whereyou are in the Administrator application. Depending on the page displayed,sometimes you can click on the bread crumb links to navigate to a differentpage.



A dark blue (shaded) tab signifies the active page. Click a light blue tab togo to that page. Some pages do not include a tab row.



Administrator Management

3

Configuring the Administrator

Use the Management features to configure the Data Services Administrator.

Related Topics• Adding repositories on page 32• Managing user roles on page 37• Adding Access Servers on page 39• Centralizing administration on page 40• Setting the status interval on page 41• Setting the log retention period on page 42

Adding repositoriesThe Administrator allows you to manage batch jobs, real-time jobs, andprofiling tasks. You must first add a repository connection to the Administratorso that you can view the jobs. Similarly, you must add a profiling repositoryconnection before you can view the tasks in that repository. After adding arepository connection:

• Jobs and logs (stored on a Job Server computer) appear in theAdministrator.

• From the Batch Job Configuration page, you can execute batch jobs.

• Repositories become available to Metadata Reports users.

• From the Profiler Configuration page, change parameter settings for theData Profiler.

• From the Profiler Task Status page, you can view and manage Profilertasks.

Related Topics• Connecting repositories to the Administrator on page 33• Changing repository connection allocation on page 36• Adapter considerations on page 36


Administrator Management3 Adding repositories

Connecting repositories to the Administrator

Use the List of Repositories page to connect a repository to an Administrator.

To add a local, central, or profiler repository connection to theAdministrator

1. Select Management > Repositories on the navigation tree.2. Click Add on the List of Repositories page.3. Enter the following information for the repository.

DescriptionOption

Logical name for a repository (used in the Administratoronly).Repository Name

The type of database storing your local, central, orprofiler repository. Select one of the following databasetypes:• DB2• Informix• Microsoft SQL Server• Oracle• Sybase ASE• MySQL

Database type


3Administrator ManagementAdding repositories

DescriptionOption

Select if your repository is an Oracle database that ispart of an Oracle Real Application Clusters (RAC) sys-tem. If you select this option, specify the completeconnection string that corresponds to the Oracle RACin the Connection string option. Obtain this RACconnection information from the Oracle TNSNames.orafile, starting from (DESCRIPTION= until the end of theentry:ARMSTRONG.ACCT =(DESCRIPTION =(ADDRESS_LIST=(FAILOVER = ON)(LOAD_BALANCE = ON)(ADDRESS= (PROTOCOL = TCP) (HOST = SERVER1)

(PORT = 1521))(ADDRESS= (PROTOCOL = TCP) (HOST = SERVER2)

(PORT = 1521))(ADDRESS= (PROTOCOL = TCP) (HOST = SERVER3)

(PORT = 1521)))(CONNECT_DATA=(SERVICE_NAME = acct.us.yourcompany.com)))

Oracle RAC

Select to have Microsoft SQL Server validate the loginaccount name and password using the information fromthe Windows operating system. Clear to use the existingMicrosoft SQL Server login account and password. Formore information on how to use Windows authenticationwith Microsoft SQL Server, refer to the Microsoft SQLServer documentation.

Windows authentication

Host name on which the database server is running.Machine Name

Port number of the database or data source.Database Port



DescriptionOption

This field requires additional information based on theDatabase Type you select.

Service Name/SID, Databasename, Server name, or Datasource

The user or owner name for the database or datasource.User name

The user's account password for the database or datasource.Password

4. (Optional) If you want to test the database information you have specifiedfor the repository, before attempting to register it with the Administrator,you can click Test.

5. Click Apply.

The Administrator validates repository connection information, and displaysit on the List of Repositories page.

To view the list of repositories connected to the Administrator

Select Management > Repositories.

The List of Repositories page lists the repositories that are connected to theAdministrator. The repository type column shows which type of repositoryyou created in the Repository Manager.

You can also remove a connection to a repository from this page.

Note: If you create a clean repository with the same name as a repositoryyou had previously connected to the Administrator, you must reconnect therepository. To do this, go to the List of Repositories page, click the repository'sname to open the Edit Repository page, then click Apply.


3Administrator ManagementAdding repositories

Changing repository connection allocation

The Administrator allocates four repository connections per user as a default.However, you can override the default value before starting the Administrator.

For Windows, modify the wrapper cmd_line section of LINK_DIR/ext/WebServer/conf by adding DCNX_POOL_LIMIT:wrapper.cmd_line=$ (wrapper.javabin)-Dcatalina.home=$ (wrapper.tomcat_home)-DLINK_DIR=$ (ACTAHOME) -DCNX_POOL_LIMIT=1-classpath $ (wrapper.class_path)

For UNIX, modify the catalina.sh script found in LINK_DIR /ext/WebServer/binby adding -DCNX_POOL_LIMIT=1 in the 'start' section (not the 'security' section)as follows:if [ "$1" = "start" ] ; thenif [ "$1" = "-security" ] ; thenecho "Using Security Manager"...

else"$_RUNJAVA" $JAVA_OPTS $CATALINA_OPTS \-DCNX_POOL_LIMIT="1" \-Djava.endorsed.dirs="$JAVA_

Adapter considerations

To access adapter instances, you must associate a repository with a JobServer that is:

• Installed on the same computer as the adapter instance.

• Configured with the following adapter-related properties:Support Adapterand SNMP communication check box selected and theCommunicationPort number set. Configure these properties using the Server Managerutility.

If these conditions have not been met, you will not be able to use theAdapterInstance node of the Administrator.



Managing user rolesThe Administrator allows you to add multiple user accounts with differentroles:

DescriptionRole

Provides access to all Administrator functional-ity.Administrator

Limited to managing secure central reposito-ries, this role is a subset of the Administratorrole. Multiuser administrators can:• Add and remove secure central reposito-

ries.

• Manage users and groups.

• View secure central repository reports.

Multiuser administrator

Provides access limited to options availablefrom the Status tabs. For example, a Monitorcan abort batch jobs but cannot execute orschedule them. A monitor can restart, abort orshut down an Access Server, service, adapterinstance, or client interface but cannot add orremove them.

Monitor


3Administrator ManagementManaging user roles

DescriptionRole

Limited to managing profiler repositories, thisrole is a subset of the Administrator role. Profil-er administrators can:• Define profiler repositories.

• Add and remove profiler users.

• Manage profiler tasks in any profiler repos-itory.

• Manage the Profiler configuration.

Profiler administrator

Limited to managing profiler tasks in the profilerrepository that is configured for the user.Profiler user

Has all Administrator privileges except cannotmodify repository, access, or CMS servers norupdate datastore settings.

Operator

To add users and their roles

1. Select Management > Users.2. Click Add to open the Add Users page.3. In the User name box, enter a new user ID.

User names and passwords for the Administrator do not need to matchthose for your system or repository.

4. In the Password box, enter the new password.5. In the Display Name box, enter another identifier for the user such as

the full name. If you have trouble recognizing a login name, you can usethis value to label the account.

6. In the Role list, select a user role.7. In the Status list, select a status for this account.


Administrator Management3 Managing user roles

You can select active or suspended. If you want to delete the user, go tothe User Management page.

8. In the Profiler repository list, select a profiler repository for this account.

You can assign a profiler repository to users with Administrator, ProfilerAdministrator, and Profiler User roles.

• A user with a Profiler User role is authorized to manage tasks only inthis profiler repository.

• For a user with an Administrator or Profiler Administrator role, therepository you specify in this option is the default repository for thisaccount. These administrators can also manage tasks in any profilerrepository.

9. Click Apply.

View the new user in the Users table on the User Management page.You can also edit or delete user IDs using from the User Managementpage.

Adding Access ServersThe Administrator acts as a front end for Access Servers connected to it.Use the Administrator to:

• Configure real-time jobs as real-time services.

• Configure real-time services with service providers.

• Monitor Access Servers, real-time services, and service providers.

You first must connect an Access Server to the Administrator so that youcan use the Administrator to create a real-time service from a real-time job.After a service starts, the Access Server brokers messages between externalapplications and Data Services.

When a message request comes in, the Access Server communicates withthe Job Server to get the repository data needed to run a real-time serviceand process the message. A reply comes back through the Access Serverto the message originator and the Access Server log records the event, whichyou can monitor from the Administrator.


3Administrator ManagementAdding Access Servers

Use the List of Access Servers page to connect an Administrator to arepository.

To connect an Access Server to the Administrator

1. Select Access Servers from the Management menu.2. Click Add.3. Enter the following information.

DescriptionOption

Host name of the computer on which the AccessServer is installed.Machine name

Port assigned to this Access Server in the ServerManager utility.Communication Port

4. (Optional) Before attempting to register the Access Server with theAdministrator, click Ping to see if the Access Server is available andexists on the computer and port you specified.

5. Click Apply.

The Administrator registers the Access Server, validates the AccessServer connection information, and shows that information on the List ofAccess Servers page.

To view a list of Access Servers connected to the Administrator, selectAccess Servers from the Management menu.

The List of Access Servers page lists the Access Servers that are connectedto the Administrator. You can also remove a connection to an Access Serverfrom this page.

Centralizing administrationYou can connect any number of repositories and Access Servers to anAdministrator, which allows you to administrate all jobs from a single, centrallocation.


Administrator Management3 Centralizing administration

Alternatively, you can set up an Administrator to manage the jobs from anindividual developer, the test repository, or different types of production jobs(batch and real-time). You can connect repositories to one Administrator,providing convenient access to a particular set of real-time jobs (for example,a set that serves a unique function such as development). However,Administrators cannot connect to each other.

To group administration by job type

1. Configure Administrators that will process a particular type of job.For example, in your production environment you can configure oneAdministrator to process batch jobs and a different Administrator toprocess real-time jobs.

2. Connect each Administrator to the repositories that contain that type ofjob.You might want to name repositories so that you can easily see the typesof jobs stored on them.

3. Connect Access Servers to any Administrators that process or managereal-time jobs.

Setting the status intervalUse the Status Interval page to specify the time period for which theAdministrator displays the status (using the red, yellow, and green statusicons) on the Batch Job Status page.

To set the status interval

1. Select Status Interval from the Management menu.2. On the Status Interval page, specify the time period.

You can filter the information on this page in three ways:

• By the last execution of each job

• By number of days

• By range of dates


3Administrator ManagementSetting the status interval

3. Click Apply.

The Administrator updates the list of job executions and the status intervaldisplays in the table title on the Batch Job Status page. The followingexample lists the jobs that have executed in the last 5 days.

Setting the log retention periodThe log retention period provides an automatic way to delete log files.

To delete log information

1. Select Log Retention Period from the Management menu.2. In the Log Retention Period box, enter the number of days you want to

retain:• Historical batch job error, trace, and monitor logs

• Current service provider trace and error logs

• Current and historical Access Server logs

The Administrator deletes all log files beyond this period. For example:


Administrator Management3 Setting the log retention period

• If you enter 1 then the Administrator displays the logs for today only.After 12:00 AM, these logs clear and the Administrator begins savinglogs for tomorrow.

• If you enter -1 Data Services will not delete logs.

• If you enter 1095 Data Services deletes logs older than approximatelythree years.

You can also delete Access Server logs manually using the Access ServerCurrent Logs and Access Server History Logs pages.

3. Click Apply.

Changes you make to the log retention period occur as a backgroundclean-up process so they do not interrupt more important messageprocessing. Therefore, you might not see logs deleted immediately whenyou select Apply. Changes can take up to an hour to take effect.

4. Choose a repository to view a list of executed batch job logs. When youselect repository name from the Batch Jobs menu, the Administratorlists the most recent job first, providing a link to each job's log.

Related Topics• Monitoring jobs on page 91

Managing database account changesData Services uses several types of user accounts and associatedpasswords. For various reasons, database account parameters such as usernames or passwords change. For example, perhaps your company'scompliance and regulations policies require periodically changing accountpasswords for security.

Related Topics• Updating local repository login parameters on page 44• Updating datastore connection parameters on page 45


3Administrator ManagementManaging database account changes

Updating local repository login parameters

If the login information, particularly the password, for a repository haschanged, Data Services provides an optional password file that all schedulesor exported execution commands use. In other words, Data Services usesthis password file to store and update connection information in one locationthat multiple schedules or exported execution commands share for thatrepository.

Note: This description does not apply to central repositories.

The password file:

• Specifies the connection information for the repository

• Can be stored in a central location for access by others who run jobs inthat repository.

• Gets created when you create or update a job schedule to minimizeassociated maintenance

Related Topics• Using a third-party scheduler on page 82

To update the repository connection information and use apassword file

1. Expand the Management node.2. Click Repositories.3. Click the repository name to configure.

The "Add/Edit Repository" page displays.

4. Edit the connection information as necessary.5. Click Apply.6. Click Generate password file to create or update the password file.

The default name and location of the file are %LINK_DIR%\conf\repositoryname.txt.


Administrator Management3 Managing database account changes

Updating job schedules

When database account information for your repository changes, the DataServices job schedules associated with that account must also be updated.When you use a password file, the job schedules access it at run time toautomatically retrieve the updated account information.

Related Topics• Scheduling jobs on page 67

Updating datastore connection parameters

If the information associated with a datastore connection changes, particularlypasswords, you can update the changes using the Administrator.

Note: Only users with Administrator role privileges can edit datastoreparameters.

To edit the connection information for an individual configurationin a datastore

1. Select Datastore Configurations from the Management menu.2. Click the configuration name to configure.3. Edit the enabled fields as necessary:4. Click Apply.

Clicking Reset returns all fields to the last set of values applied.

To edit the connection information for multiple configurationsin a datastore

1. Select Datastore Configurations from the Management menu.2. Click the datastore name to configure.

All configurations for that datastore display.

3. Edit the enabled fields as necessary.


3Administrator ManagementManaging database account changes

Click More to display the page for that individual configuration, whichincludes more options specific to it.

4. Click Apply.



Administrator Management3 Managing database account changes

Central RepositoryManagement

4

About this section

This section describes how to manage your secure central repositories usingthe Administrator.

When you create a secure central repository, the repository name appearsunder the Central Repositories node. Links under this node include:

• Users and groups — Use to add, remove, and configure users and groupsfor secure object access.

• Reports — Use to generate reports for central repository objects such asviewing the change history of an object.

Setting up users and groupsThe process for setting up users and groups is as follows:

1. Add the secure central repository to the Administrator.2. Add groups.3. Add users.4. Associate users with groups.

The following sections describes these procedures.

Note: The concept of users in the Administrator refers to setting up usersto access the Data Services Administrator application. By contrast, theUsersand Groups link under the Central Repositories node in the Administratoris for setting up rights and access to a specific central repository.

Related Topics• Managing user roles on page 37• Advanced Development Guide: Implementing Central Repository Security

To add the secure central repository to theAdministrator

1. Log in to the Administrator.2. Select Management > Central repositories.3. Add the secure central repository.


Central Repository Management4 Setting up users and groups

The repository appears on the List of Repositories page.

Related Topics• Logging in on page 23• Connecting repositories to the Administrator on page 33

To add a group to a central repository

Groups are specific to a repository and are not visible in any other local orcentral repository.

1. Expand the Central Repositories node in the navigation tree, expandthe repository to configure, and click Users and Groups.

The Groups and Users page displays.

2. On the Groups tab, click Add.3. Type a Name for the group.4. Optionally, type a Description for the group.5. Click Apply.

The group appears on the Groups tab.

To add users


2. Click the Users tab.3. Click Add.

On the Add/Edit User page, enter the following information.


4Central Repository ManagementSetting up users and groups

DescriptionOption

Type a new user name.

User names and passwords in the Adminis-trator do not need to match those for yoursystem or repository.

User name

Type a new password for the user.Password

Retype the password.Confirm password

Enter another identifier for the user such asthe full name. If you have difficulty recogniz-ing a user name, you can use this value tolabel the account.

Display name

The default group to which the user be-longs. You can change the default by select-ing another from the drop-down list.

Default group

Select a value from the drop-down list:

Active — Enables the user's account fornormal activities.

Suspended — Select to disable the loginfor that user.

Status

Optionally, type a description for the user.Description

The User is a member of list on the left shows the groups to which thisuser belongs.

4. Click Apply.


Central Repository Management4 Setting up users and groups


To add or remove a user from a group


2. Click the Group tab.3. Click the group name.4. The Member users list on the left shows the users in this group.

To add users to a group, click the user names from the Other users listand click Add users. Select multiple user names using the Ctrl or Shiftkeys.

To remove a user from the group, select a user name from the Memberusers list and click Remove users. Select multiple user names using theCtrl or Shift keys.

5. Click Apply.


Alternately, click the Users tab, click the user name, and associate the userwith one or more groups by selecting group names and adding or removingthem.

Related Topics• Advanced Development Guide: Implementing Central Repository Security

Deleting groupsYou cannot delete a group in the following instances.

• It is the default group for any user (whether or not they are active).

• It is the only group with full permissions for an object.

• A member of the group is untertaking any central repository tasks usingthe Designer.


4Central Repository ManagementDeleting groups

To delete a group


2. Click the Group tab.3. Select the check box for the group.4. Click Remove.

Viewing reportsYou can generate reports about objects in a central repository such as whichobjects a user currently has checked out or the changes made to an objectover a specified time frame.

Expand the central repository to view and expand the Reports link.

Related Topics• Object state report on page 52• Change report on page 53

Object state report

Use the object state report to view details on one or more objects such aswhether the objects are checked out and by whom.

Click theObject State Report link to display a search page with the followingcriteria (all fields are optional):

• Object name — Type an object name. You can use the % symbol as awildcard.

• Object type — For example select Batch job, Table, or Storedprocedure

• State — For example select Checked out

• User — Select a central repository user name

Click Search to generate the report. The report has the following columns.


Central Repository Management4 Viewing reports

• Object name

• Object type

• State

• User name — User account associated with the check-out or check-in

• Associated repository — The repository to which the object belongs

• Time — Check-out or check-in date and time

• Comments — Comments added when user checked out or checked inthe object

Click the object name to display the object's history.

Related Topics• Advanced Development Guide: Viewing object history

Change report

Use the change report to view the change history for an object over aspecified period of time.

Click the Change Report link to display a search page with the followingcriteria:

• Start date — Enter a date or click the calendar icon to select a start date.

• End date — Enter a date or click the calendar icon to select an end date.

• Object type — Optionally select an object type; for example batch job,table, or stored procedure.

• State — Optionally select an object state; for example Checked out

• User — Optionally select a central repository user name

Click Search to generate the report. The report has following columns.

• Object name

• Object type

• State

• Version — The version number of the object


4Central Repository ManagementViewing reports

• User name — User account associated with the check-out or check-in

• Associated repository — The repository to which the object belongs

• Time — Check-out or check-in date and time

• Comments — Comments added when user checked out or checked inthe object


Central Repository Management4 Viewing reports

Server groups

5

About this section

Use the Administrator to create and maintain server groups.

This section describes how to work with server groups.

Related Topics• Server group architecture on page 56• To add a server group on page 60• Editing and removing a server group on page 61• Monitoring Job Server status in a server group on page 63• Executing jobs using server groups on page 64

Server group architectureYou can group Job Servers on different computers into a logical Data Servicescomponent called a server group. A server group automatically measuresresource availability on each Job Server in the group and distributesscheduled batch jobs to the Job Server with the lightest load at runtime.

There are two rules for creating server groups:

• All the Job Servers in an individual server group must be associated withthe same repository, which must be defined as a default repository. TheJob Servers in the server group must also have:

• Identical Data Services versions

• Identical database server versions

• Identical locale

• Each computer can only contribute one Job Server to a server group


Server groups5 Server group architecture

.The requirement that all Job Servers in a server group be associated withthe same repository simply allows you to more easily track which jobs areassociated with a server group. Business Objects recommends that you usea naming convention for server groups that includes the name of therepository. For example, for a repository called DEV, a server group mightbe called SG_DEV.

On startup, all Job Servers check the repository to find out if they must startas part of a server group.

Compared to normal Job Servers, Job Servers in a server group each:

• Collect a list of other Job Servers in their server group

• Collect system load statistics every 60 seconds:

• Number of CPUs (on startup only)

• Average CPU load

• Available virtual memory

• Service requests for system load statistics

• Accept server group execution requests


5Server groupsServer group architecture

Load balance index

All Job Servers in a server group collect and consolidate system load statisticsand convert them into a load balance index value for each Job Server. A JobServer's load balance index value allows Data Services to normalize statisticstaken from different platforms. The Job Server with the lowest index valueis selected to execute the current job. Data Services polls all Job Servercomputers every 60 seconds to refresh the load balance index.

Job execution

After you create a server group, you can select a server group to execute ajob from the Designer's Execution Properties window or from theAdministrator's Execute Batch Job, Schedule Batch Job, and Export BatchJob pages.

When you execute a job using a server group, the server group executesthe job on the Job Server in the group that is running on the computer thathas the lightest load. The Administrator will also resynchronize a Job Serverwith its repository if there are changes made to the server group configurationsettings.

You can execute parts of your job on different Job Servers in a server group.You can select the following distribution levels from the Designer's ExecutionProperties window or from the Administrator's Execute Batch Job, ScheduleBatch Job, and Export Batch Job pages:

• Job level - A job can execute on an available Job Server.

• Data flow level - Each data flow within a job can execute on an availableJob Server.

• Sub data flow level - A resource-intensive operation (such as a sort, tablecomparison, or table lookup) within a data flow can execute on an availableJob Server.

Related Topics• Performance Optimization Guide: Using grid computing to distribute dataflows execution



Job launcher

The Job Launcher, exported as part of a job's execution commands, includesa specific command line option for server groups. You can use this optionto change the Job Servers in a server group.

Related Topics• Data Services job launcher on page 88

Working with server groups and Designer options

Some Designer options assume paths are relative to a Job Server. If yourJob Servers are on different machines from your Designer (typically the casein a production environment) you must ensure that connections and directorypaths point to the Job Server host that will run the job. Such options include:

• Source and target directories for files• Bulk load directories• Source and target connection strings to databases• Path to repositories

When using server groups consider the additional layer of complexity forconnections. For example, if you have three Job Servers in a server group:

• Use the same directory structure across your three host computers forsource and target file operations and use relative paths for file names.

• Use the same connection strings to your databases for all three Job Serverhosts.

If you use job distribution levels, the Job Servers in the server group musthave:

• Identical Data Services versions• Identical database server versions• dentical locale

• Identical operating systems

Thoroughly test the Job Server job options when working with server groups.

Adding a server group:


5Server groupsServer group architecture

• In the Administrator, use the Server Groups node to create and add aserver group.

To add a server group

1. Select Server Groups > All Server Groups.

2. Click the Server Group Configuration tab.3. Click Add.4. Follow the instructions on the Add Server Group page to create a server

group.• When you select a repository, all Job Servers registered with that

repository display. You can create one server group per repository.

• Notice that the Administrator provides a default server group name.It is the name of your repository with the prefix SG_ (for server group).You can change the default name, however, labeling a server groupwith the repository name is recommended.

• One Job Server on a computer can be added to a server group. Usethe Host and Port column to verify that the Job Servers you selectare each installed on a different host.

• When you select a repository, all Job Servers registered with thatrepository display. You can create one server group per repository.

• Notice that the Administrator provides a default server group name.It is the name of your repository with the prefix SG_ (for server group).



You can change the default name, however, labeling a server groupwith the repository name is recommended.

• One Job Server on a computer can be added to a server group. Usethe Host and Port column to verify that the Job Servers you selectare each installed on a different host.

5. After you select the Job Servers for a server group, click Apply.

The display returns to the Server Group Configuration page.

Related Topics• Monitoring Job Server status in a server group on page 63

Editing and removing a server groupYou can select a new set of Job Servers for an existing server group orremove a server group.

Trace messages are written for a change in Job Server status when youcreate, edit, or remove server groups.


5Server groupsEditing and removing a server group

• When a Job Server is upgraded to membership in a server group, thetrace message is:Collecting system load statistics, maintaining list of JobServer(s) for this server group, and accepting Job Serverexecution requests.

• When a Job Server is downgraded out of a server group, the tracemessage is:Deleting current system load statistics, and not collectingmore. Not accepting job execution requests from a servergroup.

To edit a server group

1. In the Server Group Status page, click the Configuration tab.2. In the Server Group Configuration page, click the server group that you

want to edit.3. In the Edit Server Group page, select a new set of Job Servers.4. Click Apply.

Your edited server group is saved and the display returns to the ServerGroups Configuration page.

To remove a server group

1. In the Server Group Status page, click the Configuration tab.2. In the Server Group Configuration page, select the check box for a the

server group(s) that you want to remove.3. Click Remove.

The selected server group is removed as shown in the display.

Note: If you delete Job Servers from a repository, so as to delete all the JobServers in a server group, the Administrator displays an invalid status forthe server group.


Server groups5 Editing and removing a server group

Monitoring Job Server status in a servergroup

If Job Servers are in a server group, you can view their status in theAdministrator.

• To monitor the status of these Job Servers, select Server Groups > AllServer Groups.

The Server Group Status page opens. All existing server groups aredisplayed with the Job Servers they contain.

DescriptionIndicator

A green indicator signifies that a Job Server is running.

A yellow indicator signifies that a Job Server is not running.

A red indicator signifies that the Job Server cannot connectto the repository.

If a server group contains Job Servers with a mix of green, yellow, or redindicators, then its indicator appears yellow:

Otherwise, a server group indicator displays the same color indicator asits Job Servers.

• To view the status for a single server group, select its name.


5Server groupsMonitoring Job Server status in a server group

Executing jobs using server groupsAfter you create a server group, you can select a server group to execute ajob from the Designer's Execution Properties window or from theAdministrator's Execute Batch Job and Schedule Batch Job pages.

Related Topics• Batch Jobs on page 65


Server groups5 Executing jobs using server groups

Batch Jobs

6

About this section

This section describes how to execute, schedule, and monitor batch jobsfrom the Administrator.

Before you can manage batch jobs with the Administrator, add repositoryconnections.

Related Topics• Executing batch jobs on page 66• Scheduling jobs on page 67• Monitoring jobs on page 91• Adding repositories on page 32

Executing batch jobsYou can execute batch jobs from the Administrator if their repositories areconnected to the Administrator.

To execute a job

1. Select Batch > repository.

The Administrator opens the Batch Job Status page, which lists all thejobs in the repository you just selected.

To view jobs in all repositories from this page, select Batch > AllRepositories. (The All Repositories option appears under the Batch Jobnode if more than one repository is connected to the Administrator.)

2. Click the Batch Job Configuration tab.3. To the right of the job you want to run, click Execute.

The Administrator opens the Execute Batch Job page.

4. Under Execution options, set the parameters for the execution of thisjob.

5. Under Trace options, set the trace properties for this execution of thejob.

6. Click Execute to run the job.


Batch Jobs6 Executing batch jobs

The Administrator returns to the Batch Job Status page.

Related Topics• Reference Guide: Data Services Objects, Batch Job, Parameters• Reference Guide: Data Services Objects, Batch Job, Trace properties

Scheduling jobsThere are three ways to manage job schedules.

Related Topics• Using the Data Services job scheduler on page 67• Scheduling jobs in BusinessObjects Enterprise on page 79• Using a third-party scheduler on page 82

Using the Data Services job scheduler

When you schedule batch jobs using the Data Services job scheduler, itcreates an entry in the operating system's scheduling utility on the Job Servercomputer. Windows uses the Task Scheduler and UNIX systems use theCRON utility. (Note that if you make changes to a schedule directly throughthese utilities, the Data Services job scheduler will not reflect those changes.)

Related Topics• Adding a job schedule on page 67• Activating or deactivating job schedules on page 76• Updating a job schedule on page 78• Removing a job schedule on page 78• Migration considerations on page 79

Adding a job schedule

To add a job schedule1. Select Batch > repository2. Click the Batch Job Configuration tab.3. For the job to configure, click Add Schedule.


6Batch JobsScheduling jobs

4. On the Schedule Batch Job page, enter the desired options

DescriptionOption

Enter a job schedule

Enter a unique name that describesthis schedule.

Note: You cannot rename a sched-ule after creating it.

Schedule name

Select this box to enable (activate)this schedule; then click Apply.This option allows you to createseveral schedules for a job andthen activate the one(s) you wantto run.

Active

Select a scheduler

Select where to schedule the job:

Data Services scheduler—Cre-ates the schedule on the Job Serv-er computer

BOE scheduler—Creates theschedule on the selected centralmanagement server (CMS)

Data Services scheduler

or

BOE scheduler

Select scheduled day(s) for executing the job


Batch Jobs6 Scheduling jobs

DescriptionOption

From the drop-down list on the cal-endar, select:• Day of Week to schedule the

job by the day of the week. Youcan select one or more days.Click again to deselect.

• Day of Month to schedule thejob by date. You can select oneor more dates. Click again todeselect.

If Recurring is selected, thenthe Administrator schedules thisjob to repeat every week ormonth on the selected day. Notethat if you select multiple daysof the week or month, the jobwill run on a recurring basis bydefault.

Calendar

Select scheduled time for executing the jobs



DescriptionOption

Select the job execution frequency:

Only once a day—Enter the timefor the scheduler to start the job(hours, minutes, and either AM orPM).

Multiple times a day:• For the Data Services sched-

uler, enter the time (hours, min-utes, and either AM or PM) forthe scheduler to repeatedly runthe job for the selected duration(in minutes) at the selected inter-val (in minutes).

• For the BOE scheduler, enter(in minutes) the repeat intervalto run the job. You must alsoselect all days in the calendar(for weekly or monthly).

Select a time when all of the re-quired resources are available.Typically, you want to schedulejobs to ensure they finish beforethe target database or datawarehouse must be available tomeet increased demand.

Once a day at

or

Multiple times a day

Select job execution parameters



DescriptionOption

Select the system configuration touse when executing this job. Asystem configuration defines a setof datastore configurations, whichdefine the datastore connections.

For more information, see "Creatingand managing multiple datastoreconfigurations" in theData ServicesDesigner Guide.

If a system configuration is notspecified, Data Services uses thedefault datastore configuration foreach datastore.

This option is a run-time property.This option is only available if thereare system configurations definedin the repository.

System configuration

Select the Job Server or a servergroup to execute this schedule.Job Server or server group

Select to create or update thepassword file that the job scheduleaccesses for current repositoryconnection information. Clear togenerate the batch file with a hard-coded repository information.

Use password file



DescriptionOption

Clear this check box if you do notwant to collect audit statistics forthis specific job execution. (Thedefault is selected.)

For more information about audit-ing, see “Using Auditing” in the Da-ta Services Designer Guide.

Enable auditing

Select this check box if you do notwant to collect data validationstatistics for any validation trans-forms in this job. (The default iscleared.)

Disable data validation statisticscollection

Select this check box to enable theRecovery mode when this job runs.Enable recovery

Select this check box if an execu-tion of this job has failed and youwant to enable the Recovery mode.

Recover from last failed execution

Select this check box if you want tocollect statistics that the Data Ser-vices optimizer will use to choosean optimal cache type (in-memoryor pageable). This option is not se-lected by default.

See “Using statistics for cache self-tuning” in the Data Services Perfor-mance Optimization Guide.

Collect statistics for optimization



DescriptionOption

Select this check box if you want todisplay cache statistics in the Per-formance Monitor in the Administra-tor. (The default is cleared.)

See “Monitoring and tuning cachetypes” in the Data Services Perfor-mance Optimization Guide.

Collect statistics for monitoring

Select this check box if you wantthe Data Services optimizer to usethe cache statistics collected on aprevious execution of the job. (Thedefault is selected.)

For more information, see “Usingstatistics for cache self-tuning” inthe Data Services PerformanceOptimization Guide.

Use collected statistics



DescriptionOption

Select the level within a job that youwant to distribute to multiple jobservers for processing:• Job—The whole job will execute

on an available Job Server.

• Data flow—Each data flowwithin the job can execute on anavailable Job Server.

• Sub data flow—Each sub dataflow (can be a separate trans-form or function) within a dataflow can execute on an availableJob Server.

For more information, see “Us-ing grid computing to distributedata flows execution” in the Da-ta Services Performance Opti-mization Guide.

Distribution level

5. Click Apply. Clicking Reset returns all fields to the last set of valuesapplied.



Related Topics• Designer Guide: Datastores, Creating and managing multiple datastoreconfigurations



• Managing database account changes on page 43• Reference Guide: Data Services Objects, Batch Job, Parameters• Designer Guide: Data Assessment, Using Auditing• Management Console—Metadata Reports Guide: Data ValidationDashboard Reports

• Performance Optimization Guide: Using Caches, Using statistics for cacheself-tuning

• Performance Optimization Guide: Using Caches, To monitor and tunein-memory and pageable caches

• Performance Optimization Guide: Distributing Data Flow Execution, Usinggrid computing to distribute data flows execution

Activating or deactivating job schedules

In order for a job schedule to run, it must be active.

To change an existing job schedule, you must first deactivate it, make thechanges, then reactivate it.

Related Topics• Updating a job schedule on page 78

To activate or deactivate one or more job schedules1. Select Batch > repository2. Click the Repository Schedules tab.

The Repository Schedules tab lists all schedules for all jobs in therepository and you can remove, activate, or deactivate one or moreschedules.



Alternately, you can click the Batch Job Configuration tab, then for aparticular job, click the Schedules link. The Batch Job Schedules tab listsall schedules for that particular job. Here you can add, remove, activate,or deactivate one or more schedules:

The Job Server column listed next to each schedule indicates which JobServer will execute it.

If there is a server group icon in the Job Server column, this indicates theschedule will be executed by the server group, and the schedule is storedon the indicated Job Server. To see which server group is associatedwith the schedule, roll your cursor over the server group icon.

If there is CMS icon in the Job Server column, this indicates the jobschedule is managed by a CMS.

Click the System Configuration names, if configured, to open a page thatlists the datastore configurations in that system configuration.

3. On either the Repository Schedules tab or the Batch Job Schedulestab, select one or more check boxes for a schedule.

4. Click Activate (or Deactivate).



Updating a job schedule

To edit a job schedule, you must first deactivate it, make the changes, thenreactivate it.

To update a job schedule1. Select Batch > repository2. Click the Batch Job Configuration tab.3. Click the Schedules link for the desired job.4. Click the schedule name to edit.5. The Schedule Batch Job page displays.6. If the schedule is currently active, deactivate it by clearing the Active

check box and click Apply.

Note: You do not need to deactivate the schedule to update most of thejob execution parameters at the bottom of the page. Only theschedule-related parameters require deactivation in order to update them.

7. Edit the schedule parameters as required.8. To reactivate the schedule now, select the Active check box.9. Click Apply.

The status bar at the top of the page confirms that the schedule has beencreated and/or activated.

Related Topics• Adding a job schedule on page 67

Removing a job schedule

To remove a job schedule1. Select Batch > repository2. Click the Repository Schedules tab.3. Select one or more check boxes for a schedule.4. Click Remove.

The Administrator deletes the information about this job schedule.



Migration considerations

Changes made to the Job Server, such as an upgrade, do not affectschedules created in Data Services as long as:

• The new version of Data Services is installed in the same directory asthe original version (Data Services schedulers use a hard-coded path tothe Job Server).

• The new installation uses the Job Server name and port from the previousinstallation. (This occurs automatically when you install over the existingDSConfig file.)

When you export a repository via an .atl file, jobs and their schedules (createdin Data Services) automatically export as well.

You can also import a repository .atl file including jobs and their associatedschedules (previously created in Data Services) back in to Data Services.

Remember that once imported, you must reactivate job schedules to usethem. If the job schedule uses a password file, then reactivating it willautomatically generate the password file.

Related Topics• Advanced Development Guide: Importing from a File

Scheduling jobs in BusinessObjects Enterprise

If you are using BusinessObjects Enterprise and you want to manage yourData Services job schedules in that application, first create a connection toa Central Management Server (CMS), then configure the schedule to usethat server.

Related Topics• To add a CMS connection on page 80• To create a job schedule in BusinessObjects Enterprise on page 81• To remove a CMS connection on page 82



To add a CMS connection

1. Select Management > CMS Connection.2. Click Add.3. On the CMS Connections page, enter the connection information.

The parameters in the top section are the same as when logging in toBusinessObjects Central Management Console (CMC) or InfoView. Fordetails, refer to the BusinessObjects Enterprise InfoView User's Guide.

The parameters in the bottom section (User account credentials forexecuting the program) depend on how the CMS server is set up. Fordetails, refer to "Authentication and program objects" in theBusinessObjects Enterprise Administrator's Guide.

DescriptionOption

Type the computer name that hosts the Cen-tral Management Server (CMS), a colon, andthe port number.

System

Type the CMC/InfoView user name.User name

Type the CMC/InfoView user password.Password

Select the authentication type for the serverAuthentication

User account credentials for executing the program (optional)

Note: If you do not have the following option cleared in the Business ObjectsCentral Management Console, you will be required to enter user accountcredentials in order for your schedules to run:

In the CMC, select Objects tab > Objects Settings button > Programobjects tab, clear the Use Impersonation option.



DescriptionOption

The CMS computer might require operatingsystem login credentials to run the schedule.If so, type the user name (and password) forthe applicable account.

User name

The CMS computer might require operatingsystem login credentials to run the schedule.If so, type the (user name and) password forthe applicable account.

Password

4. Click Apply.

To create a job schedule in BusinessObjects Enterprise

1. Select Batch > repository2. Click the Repository Schedules tab.3. Click the schedule name to configure.4. If the schedule is currently active, deactivate it by clearing the Active

check box and click Apply.5. Edit the schedule parameters as necessary.

Note: Time-sensitive parameters reflect the time zone of the computerwhere the Administrator is installed, not where the CMS is installed.

6. Under the Select a scheduler section, select BOE scheduler.7. From the drop-down list, select a CMS name.8. To reactivate the schedule now, select the Active check box.9. Click Apply.

The status bar at the top of the page confirms that the schedule has beencreated and/or activated.



If it doesn't already exist, BusinessObjects Enterprise creates a foldercalled Data Services and stores the schedule file and a parameters file(called schedulename.txt).

For a BOE schedule with the option Use password file selected, thenData Services also creates a password file in the Data Services folder(called repositoryname.txt)

Note: When you deactivate a schedule created on a CMS, BusinessObjectsEnterprise deletes the object. Therefore, any changes made to the calendar,etc. will be lost.

To remove a CMS connection

1. Select Management > CMS Connection.2. Select the check box for the connection to remove from the administrator.3. Click Remove.

Using a third-party scheduler

When you schedule jobs using third-party software:

• The job initiates outside of Data Services.

• The job runs from an executable batch file (or shell script for UNIX),exported from Data Services.

Note: When a third-party scheduler invokes a job, the corresponding JobServer must be running.

Related Topics• Data Services job launcher on page 88

To execute a job with a third-party scheduler

1. Export the job's execution command to an executable batch file (.bat filefor Windows or .sh file for UNIX environments).

2. Ensure that the Data Services Service is running (for that job's Job Server)when the job begins to execute.



The Data Services Service automatically starts the Job Server when yourestart the computer on which you installed the Job Server.

• You can also verify whether a Job Server is running at any given timeusing the Designer. Log in to the repository that contains your job andview the Designer's status bar to verify that the Job Server connectedto this repository is running.

• You can verify whether all Job Servers in a server group are runningusing the Administrator. In the navigation tree select Server Groups> All Server Groups to view the status of server groups and the JobServers they contain.

3. Schedule the batch file from the third-party software.

Note: To stop a Data Services job launched by a third-party schedulingapplication, press CTRL+C on the application's keyboard.

Related Topics• To export a job for scheduling on page 83• Data Services job launcher on page 88

To export a job for scheduling

1. Select Batch > repository.2. Click the Batch Job Configuration tab.3. For the batch job to configure, click the Export Execution Command

link.4. On the Export Execution Command page, enter the desired options

for the batch job command file you want the Administrator to create:



DescriptionOption

The name of the batch file or script containingthe job. The third-party scheduler executesthis file. The Administrator automatically ap-pends the appropriate extension:• .sh for UNIX

• .bat for Windows

File name

Select the system configuration to use whenexecuting this job. A system configurationdefines a set of datastore configurations,which define the datastore connections.

For more information, see “Creating andmanaging multiple datastore configurations”in the Data Services Designer Guide.

If a system configuration is not specified,Data Services uses the default datastoreconfiguration for each datastore.

This option is a run-time property. This optionis only available if there are system configu-rations defined in the repository.

System configuration

Select the Job Server or a server group toexecute this schedule.

Job Server or servergroup

Clear this check box if you do not want tocollect audit statistics for this specific job ex-ecution. (The default is selected.)

Enable auditing



DescriptionOption

Select this check box if you do not want tocollect data validation statistics for any valida-tion transforms in this job. (The default iscleared.)

Disable data validationstatistics collection

Select this check box to enable the automaticrecovery feature. When enabled, Data Ser-vices saves the results from completed stepsand allows you to resume failed jobs.

See “Automatically recovering jobs” in theData Services Designer Guide for informationabout the recovery options.

Enable Recovery

Select this check box to resume a failed job.Data Services retrieves the results from anysteps that were previously executed success-fully and re-executes any other steps. Thisoption is a run-time property. This option isnot available when a job has not yet beenexecuted or when recovery mode was dis-abled during the previous run.

Recover from last failedexecution

Select to create or update a password filethat automatically updates job schedules afterchanges in database or repository parame-ters. Clear to generate the batch file with ahard-coded repository user name and pass-word.

Use password file



DescriptionOption

Select this check box if you want to collectstatistics that the Data Services optimizer willuse to choose an optimal cache type (in-memory or pageable). This option is not se-lected by default.

See “Using statistics for cache self-tuning” inthe Data Services Performance OptimizationGuide.

Collect statistics for opti-mization

Select this check box if you want to displaycache statistics in the Performance Monitorin the Administrator. (The default is cleared.)

For more information, see “Monitoring andtuning cache types” in the Data ServicesPerformance Optimization Guide.

Collect statistics for moni-toring

Select this check box if you want the DataServices optimizer to use the cache statisticscollected on a previous execution of the job.(The default is selected.)

See “Using statistics for cache self-tuning” inthe Data Services Performance OptimizationGuide.

Use collected statistics



DescriptionOption

Select the level within a job that you want todistribute to multiple job servers for process-ing:• Job—The whole job will execute on one

job server.

• Data flow—Each data flow within the jobwill execute on a separate job server.

• Sub data flow—Each sub data flow (canbe a separate transform or function) withina data flow will execute on a separate jobserver.

For more information, see “Using gridcomputing to distribute data flows execu-tion” in the Data Services PerformanceOptimization Guide.

Distribution level

5. Click Export.

The Administrator creates command files filename.txt (the default forfilename is the job name) and a batch file for the job and writes them tothe local LINK_DIR\log directory.

Note: You can relocate the password file from the LINK_DIR\conf directory,but you must edit the filename.txt file so that it refers to the new locationof the password file. Open the file in a text editor and add the relative orabsolute file path to the new location of the password file in the argument-R " repositoryname.txt".

Related Topics• Designer Guide: Datastores, Creating and managing multiple datastoreconfigurations

• Reference Guide: Data Services Objects, Batch Job, Parameters• Designer Guide: Data Assessment, Using Auditing



• Management Console—Metadata Reports Guide: Data ValidationDashboard Reports

• Managing database account changes on page 43• Performance Optimization Guide: Using Caches, Monitoring and tuningcache types

• Performance Optimization Guide: Using Caches, Using statistics for cacheself-tuning

• Performance Optimization Guide: Distributing Data Flow Execution, Usinggrid computing to distribute data flows execution

Data Services job launcher

Data Services exports job execution command files as batch files on Windowsor CRON files on UNIX. These files pass parameters and callAL_RWJobLauncher. Then, AL_RWJobLauncher executes the job, sends it tothe appropriate Data Services Job Server, and waits for the job to complete.

Caution: Do not modify the exported file without assistance from BusinessObjects Customer Assurance.

The following shows a sample Windows NT batch file created when DataServices exports a job. ROBOT is the host name of the Job Server computer.All lines after inet:ROBOT:3513 are AL_Engine arguments, notAL_RWJobLancher arguments.D:\Data Services\bin\AL_RWJobLauncher.exe"inet:ROBOT:3513""-SrepositoryServer-Uusername-Ppassword-G"b5751907_96c4_42be_a3b5_0aff44b8afc5"-r100 -T14-CTBatch -CmROBOT -CaROBOT-CjROBOT -Cp3513"

Job launcher flag values and arguments

The following table lists job launcher flags and their values.



ValueFlag

The job launcher starts the job(s) and then waits beforepassing back the job status. If -w is not specified, thelauncher exits immediately after starting a job.

-w

The time, in milliseconds, that the Job Server waits beforechecking a job's status. This is a companion argument for-w.

-t

Status or return code. 0 indicates successful completion,non-zero indicates an error condition.

Combine -w, -t, and -s to execute the job, wait for comple-tion, and return the status.

-s

Name of the engine command file (path to a file which con-tains the Command line arguments to be sent to the engine).-C

Prints AL_RWJobLauncher version number.-v

Lists the server group and Job Servers it contains using thefollowing syntax:"SvrGroupName;JobSvr1Name:JobSvr1Host:JobSvr1Port;JobSvr2Name:JobSvr2Host:JobSvr2Port";

For example:"SG_DEV;JS1:HPSVR1:3500;JS2:WINSVR4:3505";

-S

The location and name of the password file. Replaces thehard-coded repository connection values for -S, -N, -U,-P.

-R



There are two arguments that do not use flags:

• inet address—The host name and port number of the Job Server. Thestring must be in quotes. For example:"inet:HPSVR1:3500"

If you use a server group, inet addresses are automatically rewritten usingthe -S flag arguments. On execution, the first Job Server in the groupchecks with the others and the Job Server with the lightest load executesthe job.

• server log path—The fully qualified path to the location of the log files.The server log path must be in quotes. The server log path argumentdoes not appear on an exported batch job launch command file. It appearsonly when Data Integrator generates a file for an active job schedule andstores it in the following directory: LINK_DIR/Log/JobServerName/RepositoryName/JobInstanceName

You cannot manually edit server log paths.

Job launcher error codes

The job launcher also provides error codes to help debug potential problems.The error messages are:

Error messageError number

Network failure.180002

The service that will run the schedule has not start-ed.180003

LINK_DIR is not defined.180004

The trace message file could not be created.180005



Error messageError number

The error message file could not be created.180006

The GUID could not be found.

The status cannot be returned.180007

No command line arguments were found.180008

Invalid command line syntax.180009

Cannot open the command file.180010

Monitoring jobsUsing the Administrator, you can monitor job execution of any batch job ina connected repository. You can monitor jobs that you run from theAdministrator or from the Designer.

This section discusses how you can use the Administrator to view a batchjob's overall status and statistics.

Related Topics• Overall status on page 91• Statistics on page 93

Overall status

The Batch Job Status page lists each batch job execution. Use this list toview the overall status of each execution and to access more detailedstatistics and log files.


6Batch JobsMonitoring jobs

To view overall status of executed jobs

1. Select Batch > repository.

The Batch Job Status page shows each instance of job execution on theselected repository. The list shows jobs that ran during the time periodspecified in the table title.

2. Find the overall status of a batch job execution by examining the indicatorin the Status column.


A green indicator means the batch job ran withouterror.

A red indicator means the batch job experienced anerror.

Check the End Time column to see if or when the job completed.

3. If a batch job execution has a red status, examine the trace, monitor, anderror logs for more information.

4. To view detailed information about a particular job execution, look at thedata on the Batch Job Status page.

If the job includes a server group icon in the Job Server column, thisindicates that the job was executed by a server group. You can roll yourcursor over the server group icon to view the name of the server group.The Job Server listed is the Job Server in the server group that executedthe job.


Batch Jobs6 Monitoring jobs

Note: All jobs can be executed by an explicitly selected Job Server orby a server group. If you choose to execute a job using a server group,you can use this page to see which Job Server actually executed the job.If you explicitly select a Job Server to execute a job, then even if it is alsopart of a server group, the server group icon does not appear for the jobin the Job Server column on this page.

Related Topics• Setting the status interval on page 41

Statistics

For each job execution, the Administrator shows statistics. Statistics quantifythe activities of the components of the job. You can view the following typesof statistics:

• Job statistics such as time spent in a given component of a job and thenumber of data rows that streamed through the component.

• Data flow object statistics such as the cache size used by a transformwithin a data flow.

Job statistics

To help tune the performance of a job, review job statistics.



To view job statistics1. Select Batch > repository2. On the Batch Job Status page, find a job execution instance.

Identify an instance using the page sub-title (which provides the name ofthe repository on which Data Services stores the job) and the followingcolumn headings on this page:

See Overall Status.Status

Name you gave the job in the DesignerJob Name

Name of a set of datastore configurations thatthe job uses to connect to source and targetdatabases when it executes. Each value inthis column is a link. Click the link to view theset of datastore configurations in the systemconfiguration. To change the system configu-ration, click the Configuration tab, then usethe Execute, Add Schedule or Export Exe-cution Command pages.

System Configuration

Server that ran this jobJob Server

Date and time this instance startedStart Time

Date and time this instance stoppedEnd Time

Time (in seconds) the job took to completeDuration

Times this instance ran before completingRun #

3. Under Job Information for an instance, click Monitor.



The Administrator opens the Job Server Monitor Log Viewer page. Thispage shows several statistics about this instance of job execution startingwith the name of the monitor log file.

After the file name, each line in the log provides the following information:

• Path Name — Indicates which object (step in a data flow) is executing.

• State — Indicates the run time order of the processes in the executionof the transform object and the states of each process. These are noterror status states. However, if a process state is Proceed and it neverchanges to Stop, this indicates the process ran with errors.

• Initializing — Job is initializing

• Optimizing — Job is optimizing

• Proceed — Process is executing

• Stop — Process ends without error

• Row Count — Indicates the number of rows processed through thisobject. This value updates based on the Monitor sample rate (# ofrows) set as an execution option on the Execute Batch Job page.



• Elapsed Time — Indicates the time (in seconds) since this objectreceived its first row of data.

• Absolute time — Indicates the time (in seconds) since the executionof this entire data flow (including all of the transforms) began.

Related Topics• Overall status on page 91

Data flow statistics

To help tune the performance of a data flow, review data flow statistics.

Related Topics• PerformanceOptimizationGuide: Measuring performance of Data Servicesjobs

Ignore error status

The Batch Job Status page includes an option to Ignore Error Status (buttonat end of page). Use this option if you are working through jobs with warningsor errors on this page and you want to mark a row so that you know you arefinished looking at its logs.

To ignore error status

1. Select the job or jobs that you want to ignore.2. Click the Ignore Error Status button.

The page refreshes and the rows you selected now display a green statusicon.

Deleting batch job history data

The Batch Job Status page includes an option to delete information abouthow a job ran. If you want to manually delete rows from this page, select the



rows that you want to delete, then select Delete. You can also manage thisinformation by setting the Administrator's log retention period.

Note: When you delete this job information, it also clears data validationstatistics from Data Validation Metadata Reports.

Stopping a running job

The Batch Job Status page includes an option to abort batch jobs. If a batchjob is running and you need to stop it, select the check box before the jobname and click Abort.

Trace, monitor, and error logs

You can view and delete trace, monitor, and error logs for job instances fromthe "Batch Job Status" page. The corresponding Job Server must be up andrunning to view or delete these logs.

You can set trace log options on the "Execute Batch Job" page.

You can use the Delete button on the "Batch Job Status" page to delete aset of batch log history files from a Job Server computer and its correspondingrepository.

Related Topics• Batch job logs on page 191• Statistics on page 93• Reference Guide: Data Services Objects, Log

To delete trace, monitor, and error logs for a batch job

1. Select Batch > repository.2. Select the job or jobs for which you want to delete logs.

Alternately, you can click Select All.

3. Click Delete.



Real-Time Jobs

7

About this section

This section describes how to support real-time jobs using the Administrator.

Before configuring services, add real-time job repository and Access Serverconnections to the Administrator.

Related Topics• Supporting real-time jobs on page 100• Configuring and monitoring real-time services on page 103• Creating and monitoring client interfaces on page 115

Supporting real-time jobsThe Access Server manages real-time communication between Data Servicesand external applications (such as ERP or web applications). The AccessServer determines how to process incoming and outgoing messages basedon the settings you choose for each real-time job in the Administrator.

In particular you use the Administrator to define:

• Services — A service is a name that identifies a task. The Access Serverreceives requests for a service. You associate a service with a real-timejob. The real-time job contains the real-time processing loop that canprocess requests for this service and generate a response.

• Service providers — A service provider is the computer process thatperforms a service; the service provider completes the tasks in a real-timejob. A service provider is controlled by a Job Server. A Job Server cancontrol several service providers—each service provider is a uniqueprocess or instance.

The Access Server uses services and service providers to process messagerequests.

• For example, suppose an external application sends a request to theAccess Server.

• The Access Server determines the appropriate service for the request.• Next, the Access Server finds the associated service providers and

dispatches the request to the next available service provider.• Under the control of a Job Server, that service provider completes the

processing for the request. A different Job Server might control eachservice provider.


Real-Time Jobs7 Supporting real-time jobs

The Access Server manages the entire set of service providers, implementingconfiguration changes and telling the appropriate Job Servers to start andstop service providers. At a prescribed interval, the Access Server updatesservice providers, balancing loads and implementing configuration changes.To balance loads, the Access Server monitors requests for services to ensurethat no service provider is over-used or under-used. Based on the numberof requests for a service, the Access Server tells Job Servers to start or stopservice providers.

To support real-time jobs, you must:

• Create any number of Access Servers using the Server Manager utility,then add a connection to each local or remote Access Server using theManagement node in the Administrator.


7Real-Time JobsSupporting real-time jobs

• In the Real-Time node of the Administrator, create a service for eachreal-time job under each Access Server's node.

• Create one or more service providers for each service.


Real-Time Jobs7 Supporting real-time jobs

• Start the services.

• Monitor the services.

Related Topics• Creating services and service providers on page 103• Starting and stopping services on page 109• Monitoring services on page 113

Configuring and monitoring real-timeservices

To enable an Access Server to support real-time jobs, you must configureand monitor real-time services and service providers for it.

• Configure services by specifying a real-time job and other operationalparameters.

• Configure service providers by specifying a Job Server and indicating themaximum and minimum number of instances you want the Job Server tocontrol. Each service provider is a unique process or instance controlledby a Job Server.

Related Topics• Creating services and service providers on page 103• Starting and stopping services on page 109• Updating service providers on page 112• Monitoring services on page 113

Creating services and service providers

In the Administrator, you create a service that processes requests for eachreal-time job. You also create the service providers to perform that service.A service provider is the process that completes the tasks in a real-time job.

To add a service

1. Select Real-time > Access Server > Real-Time Services.


7Real-Time JobsConfiguring and monitoring real-time services

2. Click the Real-Time Services Configuration tab.3. Click Add.4. In the Service configuration section, enter information that describes

this service.

DescriptionParameter

A unique name for this service.Service name

Click Browse Jobs to view a list of all thereal-time jobs available in the repositoriesyou connected to the Administrator. Selecta job name on this page to fill the serviceconfiguration form.

Job name

Logical name for a repository (used in theAdministrator only).Repository name

A flag that indicates whether the service willwrite trace messages.

Select Enable for the job to write tracemessages.

Enable job tracing

The maximum time the Access Server waitsfor the service to register after startup (inseconds).

Startup timeout

The maximum time the Access Server waitsfor the service to process the request (inseconds).

Queuing timeout

The maximum time the Access Server waitsfor a response from the service (in seconds).Processing timeout


Real-Time Jobs7 Configuring and monitoring real-time services


The number of times the Access Server at-tempts to restart a job that fails to respond.

Processing retry countmax

The number of requests the Access Serversends to a given real-time service beforeautomatically recycling the flow.

Recycle request countmax

If configured, select the system configurationto use when executing this service.

See “Parameters” in the Data Services Ref-erence Guide.

System Configuration

A flag that indicates whether the AccessServer attempts to automatically start thisservice when the Access Server restarts.

Select Enable if you want to automaticallystart this service when the Access Serverrestarts. This is the default setting.

If you clear Enable,when the Access Serverrestarts, it does not automatically start thisservice. If you manually attempt to start adisabled service, an error message appearsin the Service's Status column.

Enable

5. Click Apply.

The Administrator updates the configuration parameters for this service.These configuration parameters apply to all providers of this service.



When you add a new service, the Administrator creates a default serviceprovider controlled by the default Job Server.

Verify that the Job Server host name and port for the new service providerare correct by clicking the Service name and viewing the Job Servers forService section.

6. If you need to change the Job Server host name or port or want to alterthe number of service providers controlled by this Job Server:a. Click the Job Server name.

The Administrator opens the Service Provider Configuration page.b. In the Job Server list, select a Job Server to control the service

provider.



c. In Min instances and Max instances enter a minimum and amaximum number of service providers you want this Job Server tocontrol for this service.

d. Select Enable to have the Access Server start the service providerscontrolled by this Job Server. Deselect Enable to configure but notstart the service providers controlled by this Job Server.

e. Click Apply.The Administrator updates the information and returns to the JobServers for Service section of the Real-Time Service Configurationpage.

If required, you can add other service providers now or use the instructionsin the next section to add them later.

You can also add more Services now by clicking the Real-Time ServicesConfiguration tab.

7. When you are ready to have the Access Server process requests, startthe service.

Related Topics• Service startup behavior on page 130• High traffic behavior on page 131• Response time controls on page 132• To start a service on page 110• Reference Guide: Data Services Objects, Batch Job, Parameters



To add a new service provider for a service

1. Select Real-time > Access Server > Real-Time Services.2. Click the Configuration tab.3. Click the name of the service for which you want to add a service provider.4. Click Job Servers for Service.5. Click Add.6. In the Job Server list, select the Job Server that will control the service

provider.7. InMin instances andMax instances , enter a minimum and a maximum

number of service provider instances you want this Job Server to control.8. Select Enable to have the Access Server start service providers controlled

by this Job Server. Deselect Enable to configure but not start the serviceproviders controlled by this Job Server.

9. Click Apply.

If the service has already started, the Access Server adds this serviceprovider to the available list when it next updates the service providers.

If the service has not yet started, the Access Server starts enabled serviceproviders when the service starts.

Related Topics• Updating service providers on page 112

To set the service provider update interval

1. Select Real-time > Access Server > Status.2. Click the Configuration tab.3. Enter the desired Provider update interval.

This is the time interval, in seconds, between service provider updates.Valid values range from 10 to 120 seconds. The default is 30 seconds.When updating service providers, the Access Server balances loads andimplements any configuration changes you have applied to a serviceprovider.

If the provider update interval is too small, performance can decreasebecause the Access Server must frequently check for events and collect



statistics. Business Objects recommends that you set the Provider updateinterval to 30 seconds. On systems with heavy loads and productionsystems with fewer start and stop events, you can increase the interval.


To change a Job Server that executes a service

1. Select Real-time > Access Server > Real-Time Services.2. Click the Configuration tab.3. Click the service name that you want to change.4. Click the Job Servers for Service link.5. In the Job Server list, select a Job Server to control the service providers.

Job Servers are defined by host name and port number.

6. Update the maximum and minimum number of service providers, ifnecessary.

7. Select Enable to start service providers controlled by the changed JobServer when you apply this configuration. Deselect Enable to configurebut not start service providers controlled by the Job Server when youapply this configuration.

8. Click Apply.

The Administrator updates the service provider configuration.

The Access Server automatically implements the changed configurationat the next service provider update.


Starting and stopping services

After creating required services and service providers, you must start them.After you start a service or service provider, Data Services ensures that itcontinues to run. You can also use the Administrator to stop a service (such



as for maintenance). Similarly, use the Administrator to remove, enable, ordisable services and service providers.

To start a service

1. Select Real-time > Access Server > Real-Time Services.2. Select the check box next to the service or services that you want to start.3. Click Start.

The Access Server starts the minimum number of service providers forthis service.

To abort or shut down a service

1. Select Real-time > Access Server > Real-Time Services.2. Select the check box next to the service or services that you want to abort

or shutdown.• Abort — Shuts down all service providers for this service without

waiting for them to complete processing. The Access Server respondsto current and new requests for this service with an error.

• Shutdown — Shuts down all service providers for this service afterthey complete processing any current requests. The Access Serverresponds to new requests for this service with an error.

3. Click Abort or Shutdown.

Related Topics• To start a service on page 110• To disable a service on page 111

To remove a service

1. Select Real-time > Access Server > Real-Time Services.2. Click the Configuration tab.3. Select the check box next to the service or services that you want to

remove.4. Click Remove.



The Administrator stops processing this service. The Access Server shutsdown each of the service providers defined for this service and removesthe service from the list.

To disable a service

1. Select Real-time > Access Server > Real-Time Services.2. Click the Configuration tab.3. Click the service that you want to disable.4. Deselect the Enable box.5. Click Apply.

This change does not have an immediate effect on the service. Instead, theservice is disabled only when the Access Server attempts to restart thisservice, such as after the Access Server restarts.

To remove, enable, or disable a service provider

1. Select Real-time > Access Server > Real-Time Services.2. Click the Configuration tab.3. Select the check box next to a Job Server if you want to change its service

providers.4. Click one of the buttons below the list of Job Servers to perform the

appropriate action:• Remove — Discontinue using the service providers controlled by the

selected Job Servers to process requests for this service. The AccessServer shuts down the service providers and removes the Job Serverfrom the list.

• Enable — Start the service providers controlled by the selected JobServers. Each Job Server starts the minimum number of serviceproviders. The Access Server now includes the selected Job Serversin the set of available service providers. If a Job Server is alreadyenabled, this choice has no effect.

5. Shut down the service providers controlled by the selected Job Servers.The Access Server finishes processing any current requests and no longerincludes the selected Job Servers in the set of service providers availableto process requests for this service.



The Administrator completes this action during the next service providerupdate.


To restart a service provider

1. Select Real-time > Access Server > Real-Time Services.2. Select the check box next to a Job Server if you want to restart its service

providers.

Note: Only select Restart if the service providers controlled by this JobServer are currently enabled. To check, select the Configuration tab andview service provider status in the Job Servers for Service section.

3. Click Restart.

The Administrator completes this action during the next service providerupdate. The Administrator shuts down any service providers controlledby the selected Job Servers and immediately restarts the minimum numberof service providers. For example, you might restart a service providerafter a computer running its Job Server reboots following a crash.

Updating service providers

At a specified provider update interval, the Access Server updates serviceproviders. When updating service providers, the Access Server balancesthe work load—starting or stopping service providers as necessary—andimplements other events that you initiated since the last update.

When balancing the work load, the Access Server checks the number ofrequests in a service queue and the minimum idle time for a service. If thenumber of requests in a service queue is greater than the number of serviceproviders started, the Access Server tries to start a new service provider.Conversely, if the minimum idle time for a service is more than 10 minutes,the Access Server will shut down a service provider. However, the numberof service providers cannot exceed the maximum number of instancesconfigured nor can it be less than the minimum number of instancesconfigured.



When implementing events that you initiated, the Access Server:

• Enables service providers

• Disables service providers

• Reconfigures service providers

• Restarts service providers

• Adds service providers

• Removes service providers

Related Topics• To set the service provider update interval on page 108

Monitoring services

Use the Administrator to monitor services. With the Administrator you can:

• View service status — From the Access Server Status page or Real-TimeService Status page, view whether a service is running or not. Based onthis information, you can begin troubleshooting problems.

• View service provider status — From the Real-Time Service Status page,click a service name to view:

• The statistics for a particular service.

• Detailed statistics about each service provider. Using this information,you can monitor and evaluate system performance.

• The status of all service providers in that service.

• View logs — The "Access Server" node provides access to current andhistorical service provider trace and error logs.

Related Topics• To view the status of services on page 114• Service statistics on page 133• Service provider statistics on page 135• To view the statistics for a service provider on page 114• To view the logs for a service provider on page 193



To view the status of services

1. Select Real-time > Access Server > Real-Time Services.

The Administrator opens the Real-time Service Status page. For eachservice, this page shows the overall status and statistics about the numberof service providers started and the number that is possible.

2. Verify that the services are working.


A green indicator means the service is operatingproperly.

A yellow indicator means that some aspect of theservice is not working, and that the Access Serveris attempting to reestablish the service using errorhandling.

A red indicator means one or more aspects of theservice is not working, and the Access Servercannot reestablish the service.

3. If a service shows a yellow or red status, click the service name to getmore information.

Related Topics• Service statistics on page 133• Troubleshooting on page 187

To view the statistics for a service provider

1. Select Real-time > Access Server > Real-Time Services.2. Click the name of the service.



This page shows the overall statistics for the service, the service providersfor the service (listed by Job Server), and the status of each serviceprovider. Start a service to see its service provider status information.

3. Under Service Provider Status Information, click the Process ID of aservice provider to view its statistics.

The Administrator opens the Service Provider Status page.

Under Service Provider Status Information, the page shows the statisticsfor this service provider.

Related Topics• Service provider statistics on page 135• To view the logs for a service provider on page 193

Creating andmonitoring client interfacesA client is an external application that communicates with Data Servicesthrough the Access Server.

There are two types of client interfaces in the Administrator:

• RFC clients

• Message broker clients

Configure RFC clients in the Administrator for real-time jobs that use SAPERP or R/3 IDocs. To support these jobs, create a remote function call (RFC)client interface and attach IDoc configuration information to it.

Data Services creates message broker client interfaces when communicationoccurs between the Access Server and an external application that usesData Services message client libraries. To monitor message statistics, viewthe message broker clients of each Access Server as needed.

This section describes configuration and monitoring for each type of client.

For more information about using the Message Client library, see the DataServices Integrator's Guide.


7Real-Time JobsCreating and monitoring client interfaces

RFC clients

Configure IDoc message sources in the Administrator as well as in DataServices Designer. Other IDoc sources and targets need only be configuredusing the Data Services Designer.

Note: Using the Administrator, create a service for your real-time job thatcontains an IDoc as a message source before you configure an RFC Client.

An RFC client uses the SAP ERP and R/3 RFC protocol to communicatewith the Access Server. An RFC client requires connection information sothat an Access Server can register to receive IDocs from an SAP ERP orR/3 application server. An RFC client can process one or more IDoc types.An RFC client specifies which service will process a particular IDoc type andwhether or not the RFC client connection can process an IDoc type in parallel.

The process of creating an RFC client interface for IDocs has two parts:

• Adding an RFC client

• Adding IDoc configurations to an existing RFC client

Configure one RFC client per Access Server. This means that you can onlyprocess IDocs from one instance of SAP ERP or R/3. To process IDocs frommore than one instance, configure more than one Access Server.

Note: SAP ERP and R/3 function modules are responsible for IDocprocessing. In Data Services, the RFC client might fail if multiple IDocs aresent from SAP ERP or R/3 and you previously set SAP ERP or R/3's packetsize to 1. Therefore:

• Do not enable the option of immediate IDoc dispatch in SAP ERP or R/3unless the volume of produced IDocs is very light (no more than one IDocper minute).

• For batch processing of IDocs, the packet size should never be smallerthan 5 or larger than 1000. The following numbers are rough estimatesfor this parameter

:


Real-Time Jobs7 Creating and monitoring client interfaces

Packet sizeIDocs per dayIDoc Processing Vol-ume

51 - 300Light

20301 - 1000Medium

80301 - 1000Heavy

800more than 5000Very Heavy

For more information, see the BusinessObjects Data Services Supplementfor SAP.

To add an RFC client

1. Select Real-time > Access Server > Client Interfaces.2. Click the Configuration tab.3. Click Add.

The Administrator opens the RFC Client Configuration page.

4. Enter the client information in the Configuration section.

All options, except the RFC program ID, SAP gateway host name, andSAP gateway service name must match SAP ERP or R/3 datastoresettings. You might want to create a single SAP ERP or R/3 datastorefor processing real-time jobs with IDoc message sources.



DescriptionField

This is the RFC Server registrationID and is used as the Program IDin the ERP or R/3 Destination Con-figuration.

RFC program ID

User name through which DataServices connects to this SAP ERPor R/3 application server. Use thesame user name used to create theSAP ERP or R/3 datastore. Youcreated this datastore to design thejobs that include this IDoc.

ERP or R/3 user name

Password for the user accountthrough which Data Services con-nects to this SAP ERP or R/3 appli-cation server.

ERP or R/3 user password

The domain name of the computerwhere the SAP ERP or R/3 applica-tion server is running.

SAP application server name

The SAP ERP or R/3 applicationclient number.ERP or R/3 client number

The SAP ERP or R/3 applicationsystem number.ERP or R/3 system number

The domain name of the computerwhere the SAP ERP or R/3 RFCgateway is located.

SAP gateway host name



DescriptionField

The TCP/IP service name for theSAP ERP or R/3 application servergateway. Typically, this value isSAPGW and the system number.

SAP gateway service name

5. Click Apply.

The Administrator adds this client definition and returns to the ClientInterface Status page.

For more information, see the BusinessObjects Data Services Supplementfor SAP.

Adding IDoc configurations to an RFC client

Once you have created an RFC client, you can list the IDoc types that youwant to receive.

To add an IDoc configuration to an RFC client

1. Select Real-time > Access Server > Client Interfaces.2. Click the Configuration tab.3. Click the name of an existing RFC client interface.

The RFC Client Configuration page opens.

4. Click the Supported IDocs link.5. Click Add.6. Enter IDoc information:

a. In the IDoc Type box, enter the IDoc type that this SAP R/3 applicationserver will send to the Access Server.

b. In the Service Name box, enter the name of the service that willprocess this IDoc.The service identifies the job that processes this IDoc.



c. If you want the Access Server to read IDocs (of this type and from thespecified SAP ERP or R/3 source) in parallel, check the ParallelProcessing check box.Real-time services that contain an IDoc message source can beprocessed one at a time or in parallel. The Parallel Processing optionallows you to increase the number of IDoc source messages processedper minute for the IDoc type specified. This option is disabled bydefault. The Parallel Processing option allows the Access Server tosend an IDoc to a service queue (where it waits for a service provider)and continue with the next IDoc without waiting for reply. The maximumnumber of outstanding IDoc requests in the queue is the number ofIDocs received or four, whichever is smaller.

Note: Where a strict IDoc processing sequence is required, do notuse the Parallel Processing option.

7. Click Apply.8. (Optional) Click Select Real-time > Access Server > Client Interfaces.9. From the Client Interface Status page, select the check box next to the

new RFC client and click Start.

The Administrator starts the RFC client. A green indicator signifies thatthe client is running. Detailed status information is provided in the Statuscolumn.

Related Topics• Configuring and monitoring real-time services on page 103

To close connections to an RFC client interface

1. Select Real-time > Access Server > Client Interfaces.2. Select the check box next to the RFC client you want to disconnect.

If you choose Shutdown, the Access Server allows the clients to finishprocessing any active requests before closing the connection. The AccessServer responds with an error to any new requests that arrive during thatinterval.

If you choose Abort, the Access Server closes the connection to the clientwithout responding to requests currently being processed.

3. Click Shutdown or Abort.



Message Broker clients

A Message Broker client uses an XML message to communicate with a DataServices Access Server.

Message Broker clients include:

• External applications

• Adapters

• Service providers

Use the Administrator to monitor Message Broker clients.

Monitoring clients

From the Client Interface Status page, you can view the overall status of allclient connections.

To view the overall status of client connections

1. Select Real-time > Access Server > Client Interfaces.2. Verify that the RFC client connections are working.




A green indicator means each client of this type has anopen connection with Access Server.

A yellow indicator means at least one client of this type isdisconnecting.

A red indicator means the Access Server could not reservethe specified port to listen for client requests.

If an RFC client interface has a red status:

a. View details in the Status column and click the name of the client toview statistics about the particular client connection with a problem.

b. If you want to restart, abort, or shut down a client interface, clickin the navigation bar. The Administrator returns to the Client InterfaceStatus page.

c. Click Start, Abort, or Shutdown.

Related Topics• Finding problems on page 189

To monitor Message Broker clients

Select Real-time > Access Server > Client Interfaces.

Under Message Broker Clients, this page lists each message broker clientthat has registered with the Access Server along with statistics for that client.

Note: The first client in this list is the Administrator. You registered with theAccess Server when you added connection information to the Administrator.

Message broker client interface information includes:



The name of the clientName

The total time that this client has been connect-ed to the Access Server.Time Connected

The length of time since the Access Serverhas received a message from this client.Last Message Received

The length of time since the Access Serverhas sent a message to this client.Last Message Sent

The number of messages the Access Serverhas received from this client.Received Messages

The number of messages that the AccessServer has sent to this client.Sent Messages



Real-Time Performance

8

About this section

This section discusses the Access Server parameters, statistics for servicesand service providers, and how to tune the performance of services andservice providers.

Related Topics• Configuring Access Server output on page 126• Service configuration parameters on page 129• Service statistics on page 133• Service provider statistics on page 135• Using statistics and service parameters on page 136

Configuring Access Server outputYou can configure the Access Server to control its operation and output suchas sending specific event information to its trace log.

Data Services installation includes a server configuration utility called theServer Manager. The Server Manager allows you to view and change thefollowing Access Server information:

DescriptionOption

The location of the configuration and log filesfor this instance of the Access Server.

Do not change this value after the initial con-figuration.

Directory

The port on this computer the Access Serveruses to communicate with the Administratorand through which you can add additionalconfiguration information to an Access Server.

Make sure that this port number is not usedby another application on this computer.

Communication Port


Real-Time Performance8 Configuring Access Server output

DescriptionOption

Command-line parameters used by the DataServices Service to start this Access Server.

For development, consider including the fol-lowing parameters:-P -T16

where -P indicates that trace messages arerecorded, and -T16 indicates that the AccessServer collects events for services and serviceproviders.

These parameters are described in the nexttable.

Parameters

An option to control the automatic start of theAccess Server when the Data Services Ser-vice starts.

Enable Access Server

To configure an Access Server

1. SelectStart > Programs >BusinessObjects XI 3.0 >BusinessObjectsData Services > Data Services Server Manager.

2. In the Data Services Server Manager, click Edit Access Server Config.

The Access Server Configuration Editor opens.

3. Click Add to configure a new Access Server or select an existing AccessServer, then click Edit to change the configuration for that Access Server.

4. Make the appropriate changes in the Access Server Properties window.5. Click OK to return to the Access Server Configuration Editor.6. Click OK to return to the Server Manager.7. Click Restart to stop and start the Data Services Service with the new

Access Server configuration.


8Real-Time PerformanceConfiguring Access Server output

The following parameters are available to control the operation and outputof an Access Server:


Specifies the communication port for an Access Server.The default value is -A4000.-A

Disables display output.-C

Prints the parameter list to the console.-H

Enables trace messages to the console and log.-P

Indicates the location of the Access Server directory.-Rroot_directory

Determines the type of tracing information displayed inthe console and the Access Server log. Use any valueor any combination of values:

-Tvalue

system1

real-time service flow2

client4

transaction8

service16

administration64


Real-Time Performance8 Configuring Access Server output


request128

For example, to enable tracing for both system-level andservice-level operations, include the value 17 after the Tparameter.

Displays the version number of the Access Server.-V

Displays communication protocol and version number.-VC

Validates the Access Server configuration withoutlaunching the Access Server.-X

The -A and -R parameters can also be set using the Server Manager.

The -P and -T parameters can be set using the Administrator. SelectReal-Time > Access Server > Logs-Current then click the Configurationtab.

Service configuration parametersEach service contains configuration parameters that control how the AccessServer dispatches requests to the assigned real-time job. These parametersdetermine how the system handles errors that occur during operation.

Often, requirements during development differ from requirements duringproduction. Therefore, the values of your configuration parameters differduring development and production. To ensure that the system works asexpected, test the values before committing the Access Server configurationto production use.

Parameters control different categories of Access Server operation:


8Real-Time PerformanceService configuration parameters

Related Topics• To add a service on page 103• Service startup behavior on page 130• High traffic behavior on page 131• Response time controls on page 132

Service startup behavior

Use two parameters to configure how the Access Server starts serviceproviders associated with a particular service:

• Startup timeout — The maximum time the Access Server waits for a flow(service and its providers) to register after startup.

• Recycle request count max — The number of requests the Access Serversends to a given flow before automatically recycling.

When the Access Server starts, it immediately starts the service providersfor each service. If you want the Access Server to start more than oneinstance of a service to process a particular type of message, you mustdefine more than one service provider for the service.

The Job Servers launch the jobs, which in turn initiate their correspondingreal-time services. The first operation of each real-time service is to registerwith the Access Server.

If an error occurs and a real-time service fails to register, the Access Serverinstructs the Job Server to restart the job. The Access Server waits the lengthof time that you configure as the Startup timeout before instructing the JobServer to start the job again. The startup timeout is in seconds. The AccessServer continues to instruct the Job Server to restart the job until the real-timeservice registers.


Real-Time Performance8 Service configuration parameters

You can also control how many requests a particular service providerprocesses. After a provider processes the number of requests specified byRecycle request count max , the Access Server automatically recycles theservice provider—that is, the Access Server automatically stops the currentinstance of the real-time service and starts a new instance of that service.Setting this parameter to a higher value increases the time that the serviceprovider is available to accept requests for processing. Setting this parameterto a lower value refreshes any data cached in the real-time service moreoften.

High traffic behavior

Use the Queuing timeout parameter to specify the maximum amount of timethe client application must wait for a request to be processed.

If the number of requests the Access Server receives for a particular serviceexceeds the number of registered service providers that can process thoserequests, the Access Server queues the requests in the order they arereceived. When a service provider completes processing a request andresponds to the Access Server, the Access Server dispatches the nextrequest in the queue for that service to the open service provider.


8Real-Time PerformanceService configuration parameters

If there are many requests and the queue causes requests to exceed thequeuing timeout, the Access Server removes the oldest request from thequeue and responds to the client with an error indicating that the requestfailed. You can use the queueing timeout to ensure that the client receivesa timely response, even during high-traffic periods.

The queuing timeout is in seconds.

A service experiences high traffic when the available resources cannotprocess the received requests efficiently. High traffic occurs when the timemessages wait to be processed exceeds the time required to process them.

Related Topics• Using statistics and service parameters on page 136

Response time controls

Use two parameters to configure how long the Access Server waits forresponses from service providers for a particular service:

• Processing timeout

• Processing retry count max

After the Access Server sends a request to a service provider to process,the Access Server waits for the response. If the response does not arrivewithin the specified processing timeout, the Access Server sends the requestto another waiting service provider. The Processing timeout is in seconds.


Real-Time Performance8 Service configuration parameters

If the first attempt fails, the Access Server will attempt to process the requestas many times as you specify in the Processing retry countmax parameter.

If Processing retry count max is set to zero, the maximum response timeis equal to the queuing timeout plus the processing timeout.

Service statisticsThe Real-time Service Status page for a particular service shows overallstatistics.

• Number of processed requests

The number of requests for this service from any client that the AccessServer received and responded to since the last time the Access Serverstarted.

• Number of requests in the queue

The number of messages the Access Server has received from a clientfor this service but has not sent to a service provider for processing.

This value reflects the current state of the Access Server.


8Real-Time PerformanceService statistics

• Max queuing time (milliseconds)

The maximum time any request for this service waited after the AccessServer received the message and before the Access Server sent therequest to a service provider for processing.

• Average queuing time (milliseconds)

The average time that requests for this service waited after the AccessServer received the request and before the Access Server sent the requestto a service provider for processing.

• Queuing timeouts

The number of requests to which the Access Server replied to the clientwith an error indicating that there was no service provider available toprocess the request.

• Max processing time (milliseconds)

The maximum time required to process a request for this service. It is thedifference between the time the Access Server sent the request to aservice provider and the time that the Access Server responded to theclient. The processing time does not include time the request spent in aqueue waiting to be sent to a service provider.

• Average processing time (milliseconds)

The average time required to process a request for this service. It is thedifference between the time the Access Server sent the request to aservice provider and the time that the Access Server responded to theclient. The processing time does not include time the request spent in aqueue waiting to be sent to a service provider.

• Processing timeouts

The number of requests that the Access Server sent to a service providerand did not receive a response before exceeding the processing timeout.These requests are either successfully processed by another serviceprovider, or if they are left unprocessed beyond the time indicated by thequeuing timeout parameter, the Access Server returns an error to theclient.


Real-Time Performance8 Service statistics

Service provider statisticsThe Service Provider Status page shows the statistics for an instance of areal-time service.

When Data Services measures a statistic "from the start," the value doesnot restart when the Access Server restarts the service provider. The valuerestarts when the Access Server restarts.

When Data Services measures a statistic "for the current service provider,"the value restarts when the Access Server restarts the service provider,either due to error or when the service provider reaches the maximum numberof requests defined for the service.

• Max processing time (milliseconds)

The longest time it took between when the Access Server sent a messageto this service provider and when the service provider returned a response.

• Average processing time (milliseconds)

The average time it took between when the Access Server sent a messageto this service provider and when the service provider returned a response.

If you are running more than one service provider for this service, comparethis statistic with the same statistic from the other instances. If this instanceis significantly different, look for processing constraints on the computerwhere this instance runs.

• Processed requests (for the current service provider)

The number of requests that the Access Server sent to this serviceprovider to which the service provider responded.

• Processed requests (since start)

The number of requests that the Access Server sent to this serviceprovider to which the service provider responded.

• Error replies received from the start

The number of requests that the Access Server sent to this serviceprovider to which the service provider responded with an error.

• Communication errors encountered from the start


8Real-Time PerformanceService provider statistics

The number times that the communication link between the Access Serverand this service provider failed.

• Timeout errors encountered from the start

The number of times the Access Server sent requests to this serviceprovider and did not receive a response within the time specified by theprocessing timeout.

• Service provider connections (restarts) from the start

The number of times the Access Server restarted this service providerwhen it did not receive a response from the service provider.

• The last time of a successful flow launch

The system time when the Access Server last started the real-time serviceassociated with this service provider. If the Access Server neversuccessfully started an instance of this service provider, the value is "N/A."

This time is from the computer running the Access Server.

• Time since start attempt

The amount of time since the Access Server last attempted to start thisservice provider. This value reflects successful and unsuccessful attempts.

• Time since last request start

The amount of time since the Access Server last sent a request to thisservice provider. This value reflects successful and unsuccessful attempts.

Using statistics and service parametersYou can use the statistics for a service to tune the service parameters.

• Average and maximum processing time

If the average or maximum processing time for a service provider is equalor close to the processing timeout value resulting in processing timeouts,consider increasing the processing timeout parameter for the service.

• Maximum queuing time

In a tuned system, the maximum and average queuing times should beclose together, the difference being an indication of the traffic distribution


Real-Time Performance8 Using statistics and service parameters

for this service. Values should not approach the value of the queuingtimeout parameter listed for the service.

If the maximum queuing time for a service provider is equal or close tothe queuing timeout parameter and there are queuing timeouts listed,consider the following changes:

• Increase the queuing timeout parameter for the service

• Increase the number of service providers available, either controlledby the same Job Server host or by a different Job Server

If you find that the average time in the queue is longer than the averageprocessing time, the traffic for this service is too high for the resourcesprovided. Consider running multiple service providers to process the samemessage type. You can add the same job many times in the service list,or you can add the same job controlled by a different Job Server on aseparate computer to the service list.

If you find that the average queuing time is growing, consider increasingthe queuing timeout or adding processing resources.

• Processing timeouts

If you see processing timeouts and service providers restartingsuccessfully, consider increasing the number of processing retries allowedfor the service.


8Real-Time PerformanceUsing statistics and service parameters


Real-Time Performance8 Using statistics and service parameters

Profile Server Management

9

About this section

This section describes how to use the Administrator to manage the data inthe profiler repository and manage tasks on the profiler server.

The Data Profiler executes on a profiler server to provide the following dataprofiler information that multiple users can view:

• Column analysis—This information includes minimum value, maximumvalue, average value, minimum string length, and maximum string length.You can also generate detailed column analysis such as distinct count,distinct percent, median, median string length, pattern count, and patternpercent.

• Relationship analysis—This information identifies data mismatchesbetween any two columns for which you define a relationship, includingcolumns that have an existing primary key and foreign key relationship.

You can execute the Data Profiler on data contained in databases and flatfiles. Databases include DB2, Oracle, SQL Server, Sybase, and AttunityConnector for mainframe databases. See the Data Services Release Notesfor the complete list of sources that the Data Profiler supports.

This section assumes that you have already installed Data Services whichincludes the Data Profiler.

Related Topics• Designer Guide: Data Assessment, Using the Data Profiler

Defining the profiler repositoryThe Data Profiler repository is a set of tables that holds information aboutyour data that the Data Profiler generates.

To define a profiler repository

1. Create a database to use as your profiler repository. The profiler repositorycan be one of the following database types: DB2, MySQL, Oracle, SQLServer, or Sybase.

2. Create a profiler repository on the Repository Manager. Select Profilerin the Repository type option.


Profile Server Management9 Defining the profiler repository

3. Associate the profiler repository with a Data Services job server.

Related Topics• Installation Guide: After Installing Data Services, Using the ServerManager

• Installation Guide: After Installing Data Services, Using the RepositoryManager

Connecting repositories to theAdministrator

The Data Services Administrator manages the data in the profiler repositoryand manages tasks in the profiler server. Use the List of Repositories pageto connect an Administrator to a repository.

To add a local, central, or profiler repositoryconnection to the Administrator

1. Select Management > Repositories on the navigation tree.2. Click Add on the List of Repositories page.3. Enter the following information for the repository.

DescriptionOption

Logical name for a repository (usedin the Administrator only).Repository Name

The type of database storing yourlocal, central, or profiler repository.There are several options:• DB2• Microsoft SQL Server• MySQL• Oracle

Database type


9Profile Server ManagementConnecting repositories to the Administrator

DescriptionOption

Host name on which the databaseserver is running.Machine Name

Port number of the database or da-ta source.Database Port

This field requires additional infor-mation based on the DatabaseType you select.

Service Name/SID, Databasename, Server name, or Data source

The user or owner name for thedatabase or data source.User name

The user's account password forthe database or data source.Password

4. (Optional) If you want to test the database information you have specifiedfor the repository, before attempting to register it with the Administrator,you can click Test.

5. ClickApply. The Administrator validates repository connection information,and displays it on the List of Repositories page.

When you connect a profiler repository, the repository name appearsunder the Profiler Repositories node in the navigation tree.


Profile Server Management9 Connecting repositories to the Administrator

To view the list of repositories connected to theAdministrator

Select Management > Repositories.

The List of Repositories page lists the repositories that are connected to theAdministrator. The repository type column shows which type of repositoryyou created in the Repository Manager: Local, Central, or Profiler.

You can also remove a connection to a repository from this page.

Note: If you create a clean repository with the same name as a repositoryyou had previously connected to the Administrator, you must reconnect therepository. To do this, go to the List of Repositories page, click the repository'sname to open the Edit Repository page, then click Apply.

Defining profiler usersYou can use the default user name (admin) and password (admin) to connectto the Profiler Server. Use the following procedure to define more profilerusers.


9Profile Server ManagementDefining profiler users

To define a profiler user

1. Access the Add Users page and enter the user information:a. Select Management > Users.b. Click Add to open the Add Users page.c. Enter a new user ID and new password.d. In the Display Name box, enter another identifier for the user such

as the full name. If you have trouble recognizing a login name, youcan use this value to label the account.

2. In the Role list, select a Profiler Administrator, Profiler User, orAdministrator role.

3. In the Status list, keep the default value active.4. In the Profiler repository list, select a profiler repository for this account.

• A user with a Profiler User role is authorized to manage tasks only inthis profiler repository.

• For a user with an Administrator or Profiler Administrator role, therepository you specify in this option is the default profiler repositoryfor this account. These administrators can also manage tasks in anyprofiler repository.

5. Click Apply.


Profile Server Management9 Defining profiler users

Configuring profiler task parametersSet configuration parameters to control the amount of resources that profilertasks use to calculate and generate profiler statistics.

Note: If you plan to use Detailed profiling or Relationship profiling, ensurethat you specify a pageable cache directory that:

• Contains enough disk space for the amount of data you plan to profile.

• Is on a separate disk or file system from the Data Services system.

Related Topics• Installation Guide: Using the Server Manager, To configure run-timeresources for Job Servers


9Profile Server ManagementConfiguring profiler task parameters

To configure profiler task parameters

1. Select Management > Profiler Configuration to access the ProfilerConfiguration page.

2. Keep or change the parameters values listed on the Profiler Configurationpage.

The Profiler Configuration page groups the parameters into the followingcategories:

• Task execution parameters

• Task management parameters

Related Topics• Task execution parameters on page 146• Task management parameters on page 148

Task execution parametersThe Profiler Configuration page groups the Task execution parameters into subcategoriesReading Data, Saving Data, and Performance, as the following table shows.

DescriptionDefault valueParameterTask executionsubcategory

Maximum number of rows toprofile.

You might want to specify amaximum number of rows toprofile if the tables you pro-file are very large and youwant to reduce memory con-sumption.

AllProfiling sizeReading Data


Profile Server Management9 Configuring profiler task parameters


Profile the first row of thespecified number of sam-pling rows.

For example, if you set Pro-filing size to 1000000 andset Sampling rows to 100,the Profiler profiles rowsnumber 1, 101, 201, and soforth until 1000000 rows areprofiled. Sampling rowsthroughout the table can giveyou a more accurate repre-sentation rather than profilingjust the first 1000000 rows.

1Sampling rowsReading Data

Number of distinct values tosave in the profiler reposito-ry.

100Number of distinct val-uesSaving Data

Number of patterns to savein the profiler repository.100Number of patternsSaving Data

Number of days to keep pro-filer results in the profilerrepository.

90Number of days to keepresultsSaving Data

Number of records to savein the profiler repository foreach attribute.

100Number of records tosaveSaving Data




Number of rows to save be-fore a commit is issued.5000Rows per commitSaving Data

Number of parallel process-ing threads that the profilertask can use.

2Degree of ParallelismPerformance

Number of file processingthreads for file sources.2File processing threadsPerformance

Task management parametersThe Profiler Configuration page groups the Task management parameters into subcategoriesBasic and Advanced, as the following table shows

DescriptionDefault valueParameterTask manage-ment subcatego-ry

Maximum number of profilertasks to run simultaneously.10Maximum concurrent

tasksBasic


Profile Server Management9 Configuring profiler task parameters


Number of days that mustelapse before a profiler taskis rerun for the same table orkey columns when the userclicks the Submit option. TheSubmit option is on the Sub-mit ColumnProfile Requestand Submit RelationshipProfile Request windows inthe Data Services Designer.

Default is 0 to always rerunthe profiler task when the us-er clicks the Submit option.In other words, there is nolimit to the number of DataProfiler tasks that can be runper day.

To override this interval, usethe Update option on theProfile tab of the View Datawindow in the Data ServicesDesigner.

0Refresh interval (days)Basic

Number of seconds to sleepbefore the Data Profilerchecks for completion of aninvoked task.

Invoked tasks run syn-chronously, and the DataProfiler must check for theircompletion.

5Invoke sleep interval(seconds)Advanced




Number of seconds to sleepbefore the Data Profiler at-tempts to start pending tasks.

Pending tasks have not yetstarted because the maxi-mum number of concurrenttasks was reached.

10Submit sleep interval(seconds)Advanced

Number of minutes a profilingtask can be inactive beforethe Data Profiler cancels it.

1Inactive interval (min-utes)Advanced

Monitoring profiler tasks using theAdministrator

You can monitor your profiler task by name in either the Data ServicesDesigner or the Data Services Administrator.

On the Data Services Administrator, you can see the status of profiler tasks,cancel profiler tasks, or delete a profiler task with its generated profilestatistics.

Related Topics• Designer Guide: Data Assessment, Monitoring profiler tasks using theDesigner


Profile Server Management9 Monitoring profiler tasks using the Administrator

To monitor a profiler task in the Data ServicesAdministrator

1. Expand the Profiler Repositories node.2. Click on your profiler repository name.3. The Profiler Tasks Status window displays.

This status window contains the following columns:

DescriptionColumn

If you want to cancel a profiler taskthat is currently running, place acheck mark in this box and clickCancel.

If you want to delete a profiler taskand its profiler data from the profilerrepository, place a check mark inthis box and click Delete.

If you click Delete on a runningtask, the Profiler cancels the taskbefore it deletes the data.

Select


9Profile Server ManagementMonitoring profiler tasks using the Administrator

DescriptionColumn

The status of a profiler task can be:• Done — The task completed

successfully.

• Pending — The task is on thewait queue because the maxi-mum number of concurrenttasks has been reached or an-other task is profiling the sametable.

• Running — The task is currentlyexecuting.

• Error — The task terminatedwith an error.

Status

Name of the profiler task. The nameis a link to the Profiler Task Itemsreport (see step 4 below).

Task Name

The names of the tables on whichthe profiler task was run.Description

The identification number for thisprofiler task instance.Run #

The date and time that this profilertask last performed an action (suchas submitted or attributes calcula-tions completed for a column).

Last Update



DescriptionColumn

Blank if the profiler task completedsuccessfully. Error message if theprofiler task failed.

Status Message

4. Click the task name to display the Profiler Task Items report that displaysthe profiling type that was done for each column:

This Profiler Task Items report contains the following columns:



DescriptionColumn

The status for each column onwhich the profiler task executed.The status can be:• Done — The task completed

successfully.

• Pending — The task is on thewait queue because the maxi-mum number of concurrenttasks has been reached or an-other task is profiling the sametable.

• Running — The task is currentlyexecuting.

• Error — The task terminatedwith an error.

Status

The column number in the datasource on which this profiler taskexecuted.

Item

The machine name and port num-ber of the job server where theprofiler task executed.

Job Server

The Data Services process ID thatexecuted the profiler task.Process ID



DescriptionColumn

Indicates what type of profiling wasdone on each column. The ProfilingType can be:• Single Table Basic — Column

profile with default profile statis-tics

• Single Table Detailed — Col-umn profile with detailed profilestatistics

• Relational Basic — Relationalprofile with only key column data

• Relational Detailed — Relationalprofile with data saved from allcolumns

Profiling Type

Name of the datastoreDatastore

Name of the data source (table, flatfile, or XML file)Source

Name of the column on which theprofiler task executed.Column

The date and time that this profilertask last performed an action (suchas submitted or profile calculationscompleted for a column).

Last Update

Blank if the profiler task completedsuccessfully. Error message if theprofiler task failed.

Status Message



Adapters

10

About this section

This section describes how to add an adapter to the Data Services system,how to start an adapter instance, and how to monitor an adapter's operationinstances.

Related Topics• Overview of adapters on page 158• Adding and configuring adapter instances on page 160• Starting and stopping adapter instances on page 165• Monitoring adapter instances on page 166

Overview of adaptersA Data Services adapter is a Java program that allows Data Services tocommunicate with front-office and back-office applications. Depending onthe adapter implementation, Data Services adapter capabilities include theability to:

• Browse application metadata

• Import application metadata into the Data Services repository

• Move batch and real-time data between Data Services and informationresource applications

Data Services adapters can handle the following types of metadata: tables,documents, functions, outbound messages, and message functions. Eachof these can be used in real-time or batch jobs. Outbound messages andmessage functions are the only objects that include operations.

An adapter can process several predefined operations. An operation is aunit of work or set of tasks that the adapter completes. Operations include:

• Taking messages from an application and send them to a real-time servicefor processing, possibly returning a response to the application

• Taking messages from a real-time service and send them to an applicationfor processing, possibly returning a response to the real-time service

• Taking messages produced by a function call inside a real-time service,send the messages to an application, and return responses to the function


Adapters10 Overview of adapters

An adapter connects Data Services to a specific information resourceapplication. You can create one or more instances of an adapter. Eachadapter instance requires a configuration file. That configuration file definesthe operations available.

All adapters communicate with Data Services through a designated JobServer. You must first install an adapter on the Job Server's computer beforeyou can use the Administrator and Designer to integrate the adapter withData Services. See your specific adapter's documentation for its installationinstructions.

After installing the adapter, configure its instances and operations in theAdministrator before creating adapter datastores in the Designer becauseyou must select an adapter instance name as part of an adapter datastoreconfiguration. It might help to think of the Adapter Instances node of theAdministrator as part of your adapter datastore configuration.

To create an adapter datastore connection in theDesigner

1. Use the Server Manager utility to configure a Job Server that supportsadapters and an associated repository.

2. Use the Administrator to:• Add a connection to the Job Server's associated repository by selecting

Management > Repositories > Add.

• Add at least one instance of the adapter to the Data Services systemby selecting Adapter Instances > Job Server > Configuration >Add.

• If the adapter instance includes operations, add at least one operationfor each adapter instance.

• Start the adapter instance (operations start automatically)

3. Second, use the Designer to create an adapter datastore and importmetadata. Use the metadata accessed through the adapter to createbatch and/or real-time jobs.

When you are ready to administer jobs that include objects that use theadapter datastore, add the repository connection for the jobs to the


10AdaptersOverview of adapters

Administrator, open the Batch or Real-Time nodes, and use them as youwould for any job.

You can also monitor adapter instances and operations in the AdapterInstances node of the Administrator.

Related Topics• Adapter considerations on page 36• Installation Guide: Using the Server Manager• Designer Guide: Datastores, Adapter datastores

Addingand configuringadapter instancesUse the Administrator to add adapter instance configuration information tothe Data Services system and to edit an existing configuration.

Until you add an adapter interface using the Administrator, you cannot runjobs using information from that adapter.

To add an adapter instance

1. Select Adapter Instances > Job Server.2. Click the Configuration tab.3. Click Add.4. Click an adapter from the list of those installed on the Job Server with

which you are working.

Note: The HTTP adapter and the Web Services adapter automaticallyinstall with every Job Server. Both adapters allow you to call externalapplications from Data Services, one using the HTTP or HTTPS protocoland the other using the SOAP protocol. Use the Web Services adapterto create outbound calls because it automatically configures and startswhen a Job Server is enabled for use with adapters. However, if you wantto use the HTTP adapter you can, but you must build it like any otherData Services adapter.

For more information about the HTTP adapter see the Data ServicesHTTP Adapter Guide.

5. Enter the required information to create an adapter instance.


Adapters10 Adding and configuring adapter instances

6. Click Apply.

The Administrator adds the adapter instance to the list available to theData Services system.

Related Topics• To create an adapter datastore connection in the Designer on page 159• Support for Web Services on page 169• Adapter instance configuration information on page 161

To edit an adapter's configuration

1. Select Adapter Instances > Job Server.2. Click the Configuration tab.3. Click the name of the adapter instance that you want to edit.

The Administrator displays the current configuration information.

4. Edit the configuration information.5. Click Apply.

The Administrator updates the information.

Related Topics• Adapter instance configuration information on page 161

Adapter instance configuration information

Complete the following fields in the Administrator to set up an adapterinstance in the Data Services system:

Note: TheAdapter Instance Name is the only option required if your adapterinstance is for batch jobs.

• Adapter Instance Name

A unique name that identifies this instance of the adapter.

• Access Server Host


10AdaptersAdding and configuring adapter instances

To run an adapter instance in a real-time job, you must configure a servicethat will be called from a given Access Server. Enter the host ID of thecomputer running the Access Server for which you will configure a servicefor a real-time job that contains this adapter instance.

• Access Server Port

The communication port of the Access Server host is used to both connectan Access Server to Data Services components and to broker messageswith external applications. After you log in to the Administrator, selectReal-time > Access Server > Client Interfaces to view an AccessServer's message broker port information.

• Adapter Retry Count

Number of times to retry an adapter instance if it fails or crashes. Enter0 to indicate no retries. Enter a negative number to retry the instanceindefinitely.

• Adapter Retry Interval

Number of milliseconds to wait between adapter retry attempts.

• Classpath

All adapter Java programs require specific jar files in the CLASSPATHto use when starting the javaw.exe. For example:

• LINK_DIR\lib\acta_adapter_sdk.jar

• LINK_DIR\lib\acta_broker_client.jar

• LINK_DIR\lib\acta_tool.jar

• LINK_DIR\ext\lib\xerces.jar

Your adapter program might require different jar files.

You can change the system CLASSPATH environmental variable, or youcan use this option to enter a CLASSPATH parameter for the requiredjar files. If you use this option, enter, for example:C:\Data Services\lib\acta_adapter_sdk.jar; Data Services\lib\acta_broker_client.jar;

Data Services\lib\acta_tool.jar;Data Services\ext\lib\xerces.jar

• Autostart

Enable the adapter interface to start automatically when Data Servicesstarts by setting this option to True.



• Trace mode

A flag that controls the amount of trace messages the adapter writes.There are two settings:

• True— Select to have the adapter interface write additional informationmessages to help debug problems

• False — Select to have the adapter interface write only minimalinformation messages

The adapter writes trace messages to theadapter_instance_name_trace.txt file in the LINK_DIR\adapters\logsdirectory.

• Additional java launcher options

In addition to the classpath, you can use additional options when launchingjava processes (javaw.exe for Windows and java.exe for UNIX platforms).Here are some examples:

• If you do not define a value in this box, the default options, for memoryusage, are: -Xms128m -Xmx256m.

• If you get an out-of-memory error from an adapter, then you canre-configure its instance by editing the additional java launcher options.For example: -Xms512m -Xmx1024m

• If an adapter requires that you define a system property, do so byediting the additional java launcher options: -Xms128m -Xmx256mfoo="string"

• Adapter type name

(Read-only) The name of the adapter used to create this instance.

• Adapter version

(Read-only) The version of the adapter used to create this instance.

• Adapter Class

(Read-only) A name that identifies the adapter class. The name dependson the type of adapter:

• For prepackaged adapters, see the adapter documentation.


10AdaptersAdding and configuring adapter instances

• For custom adapters, this is the adapter's fully qualified Java classname:package_name.class_name

where:

package_name is the Java package name for the adapter as definedin the adapter's Java file.

class_name is the Java class name for the adapter as defined in theadapter's Java file.

For example, suppose the adapter's Java file contains these lines:package com.acta.adapter.SiebelAdapterpublic class SiebelAdapter implements

Then, the adapter class name is:com.acta.adapter.SiebelAdapter.SiebelAdapter

• Root Directory

Examine the adapter's root directory name. Edit this name as needed.

To add operation instances to an adapter instance

1. Select Adapter Instances > Job Server.2. Click the Configuration tab.3. Click Operations under Dependent Objects.4. Click Add to configure a new operation.

Here you can also click the link of an existing operation instance to editits configuration.

5. Select an operation type from the list.

The options that appear on this page depend on the adapter's specificdesign.

6. Click Apply.7. Complete the operation instance configuration page.

The options and descriptions that appear on this page depend on theadapter's specific design. Consult your adapter-specific documentationfor details.



8. Click Apply.

Starting and stopping adapter instancesUse the Administrator to start and stop an adapter instance and its operations.

• After you configure an adapter instance, each time you stop and start theAccess Server, you stop and start the adapter instance and its operations.

• After you restart an adapter instance, the service that uses it fails toprocess the next message it receives. Therefore, when you restart anadapter instance, also restart its associated services.

To start an adapter instance

1. Select Adapter Instances > Job Server.2. Select the check box next to the adapter instance you want to start.3. Click Start.

The Administrator starts the adapter instance and all of its operations.

To stop an adapter instance

1. Select Adapter Instances > Job Server.2. Select the check box next to the adapter instance you want to stop.3. Click either Shutdown or Abort:

• Select Shutdown to stop an adapter and all of its operations gracefully.The adapter will complete any pending operations before shuttingdown.

• Select Abort if you want to stop all operations immediately. SelectAbort only if incomplete operations are acceptable.

To start or stop an adapter operation instance

1. Select Adapter Instances > Job Server.


10AdaptersStarting and stopping adapter instances

2. Select the check box next to the operation instance you want to start orstop.

When you start an adapter instance, its operations will also start. However,you can also start and stop individual operation instances manually usingthis page.

3. Click either Start or Shutdown.

Monitoring adapter instancesUse the Administrator to monitor adapters and their operations.

To monitor the adapter instances and operations

1. Select Adapter Instances > Job Server

The Adapter Instance Status page lists each adapter instance and itsoperations.

2. Find the overall status of a particular adapter instance or operation byexamining the indicators.


A green indicator signifies that the adapter instance or op-eration has started and is currently running.

A yellow indicator signifies that the adapter instance or op-eration is not currently running.

A red indicator signifies that the adapter instance or opera-tion has experienced an error.

For each operation, this page lists four statistics:

• Requests Processed


Adapters10 Monitoring adapter instances

The number of requests for this operation instance that wereprocessed. Processing of these requests is complete.

• Requests Pending

The number of requests for this operation instance that are stillpending. Processing of these requests is not complete.

• Requests Failed

The number of requests for this operation instance that have failed.The operation has stopped processing these requests.

• Status

For operations, displays error text.

You can also find more detailed adapter instance information in theStatus column. Possible values include:

• Initialized

• Starting

• Started

• Shutting Down

• Shutdown

• Error text — Displays the last error message that occurred as theadapter instance shut down or indicates that the configuration haschanged. To allow the adapter instance to use the changes, restartthe adapter instance.

For more detailed information about the adapter instance, view the errorand trace log files.

To monitor adapter instance statistics

1. Select Adapter Instances > Job Server.2. Click the name of an adapter instance.

The statistics for the instance appear. The options and descriptions thatappear on this page depend on the adapter's specific design. Consultyour adapter-specific documentation for details.


10AdaptersMonitoring adapter instances


Adapters10 Monitoring adapter instances

Support for Web Services

11

For information about using Data Services as both a Web services serverand client, see the Data Services Integrator's Guide.


Support for Web Services11

Support for HTTP

12

About this section

The HTTP functionality is installed with every Data Services Job Server. Thissection describes how to configure and use this functionality with DataServices. Topics include:

Related Topics• Overview on page 172• Adapter installation and configuration on page 173

OverviewHypertext Transfer Protocol (HTTP) is an application-level protocol fordistributed, collaborative, hypermedia information systems. HTTP has beenin use by the World-Wide Web global information initiative since 1990 andits use has increased steadily over the years, mainly because it has provenuseful as a generic middleware protocol.

HTTP is a request/response protocol. A client sends a request to a serverspecifying a "request method", a Universal Resource Identifier (URL), andprotocol version, followed by a message containing client information andusually body content.

The server responds with a status line including the message's protocolversion and a success or error code, followed by a message containingserver information and usually body content.

HTTP communication usually takes place over TCP/IP connections. Thedefault port is TCP 80 [19], but other ports can be used. This does notpreclude HTTP from being implemented on top of any other protocol on theInternet, or on other networks. HTTP only presumes a reliable transport; anyprotocol that provides such a guarantee can be used.

HTTP can also utilize a Secure Socket Layer (SSL) to implement securityat the protocol level. In this manner, data exchange is protected from anyunscrupulous elements.

Business Objects Data Services supports HTTP in the following manner:

• Data transfer can be done using either HTTP or HTTPS (HTTP with SSL)protocols

• The transport mechanism is always TCP/IP


Support for HTTP12 Overview

• Both batch and real-time jobs can request data from HTTP-enabledservers, acting as HTTP clients

• Real-time jobs can be executed via HTTP requests, therefore makingData Services act as an HTTP server

Adapter installation and configurationThe ability to handle requests to execute real-time jobs as an HTTP servercomes pre-configured with Data Services and it is implemented as a servletdeployed inside the Data Services Web Server. The ability to call otherservices as an HTTP client is implemented as an HTTP adapter and itrequires further configuration, as explained in subsequent sections.

URL for HTTP requests to Data Services

The Data Services server URL format is:http://host:port/DataServices/servlet/HTTP?ServiceName={GetService}

Where:

• host is the IP address/host name of the Data Services Access Server• port is the port number of the Access server

These values are the same as in the URL of the Data Services Administrator.

Configuring the HTTP adapter

When you configure the HTTP adapter you must configure one or moreinstances of the adapter as well as one or more operation instances for eachadapter instance.

Adapter operations identify the integration operations available with theconfigured adapter instance.

Operations provided with the HTTP Adapter include:

• Request/Reply Operation


12Support for HTTPAdapter installation and configuration

This operation is used to execute a remote HTTP service in the RequestReply mode i.e. it makes the request to the remote machine where theHTTP server is running and wait for the reply.

• Request/Acknowledge Operation

This operation is used to execute a remote HTTP service in the RequestAcknowledge mode i.e. it makes the request to the remote machine wherethe HTTP Adapter server is running and does not wait for the reply;instead, it sends an acknowledgement if the operation is successful.

All Data Services adapters communicate with Data Services through adesignated Adapter Manager Job Server. Use the Server Manager utility toconfigure adapter connections with the Adapter Manager Job Server.

Use the Administrator to add an HTTP adapter to the Data Services systemand to edit existing adapter configurations. Until you add the adapter in theAdministrator, you cannot run jobs using information from that adapter.

Related Topics• Installation Guide: Using the Server Manager, To configure Job Servers

To add an adapter instance in the Administrator

1. Select Adapter Instances > Job Server.2. Click the Configuration tab.3. Click Add.4. Select the HTTP adapter from the list of those available on this Job Server.5. Enter the required information to create an HTTP adapter instance.6. Click Apply.

The Administrator adds the adapter instance to the list of those availableto the Data Services system.

Related Topics• Adapter instance startup configuration on page 175


Support for HTTP12 Adapter installation and configuration

Adapter instance startup configuration

Complete the following fields in the Administrator to set up an HTTP adapterinstance in the Data Services system:

• Adapter instance name — Enter a unique name that identifies thisinstance of the HTTP Adapter.

• Access Server Host — Enter the host ID of the computer running theAccess Server that connects to this adapter instance. This information isused by the web application server.

• Access Server Port — The Access Server host's message broker port.After you log into the Administrator for this Access Server, select Configuration > Interfaces to view message broker port information.

• Adapter Retry Count — The number of times for Data Services to retrythe adapter instance if it fails or crashes. Enter a negative number to retryindefinitely or enter 0 for no retries.

• Adapter Retry Interval — Sets the number of milliseconds betweenadapter retry attempts.

• Classpath — All adapter Java programs require specific jar files in theCLASSPATH to use when starting the javaw.exe. For example:

LINK_DIR/lib/acta_adapter_sdk.jarLINK_DIR/lib/acta_broker_client.jarLINK_DIR/lib/acta_tool.jarLINK_DIR/ext/lib/xerces.jarLINK_DIR/lib/acta_http_adapter.jarLINK_DIR/lib/jcert.jarLINK_DIR/lib/jnet.jarLINK_DIR/lib/jsse.jar

• Autostart — When set to True, the adapter interface automatically startswhen the Administrator starts.

• Trace mode — Set this flag to control the number of trace messages theadapter writes. There are two settings:

• True — The adapter interface writes additional information messagesto help debug problems.

• False — The adapter interface writes minimal information messages.The adapter writes trace messages to the adapter_in



stance_name_trace.txt file in the LINK_DIR/adapters/logsdirectory.

• Application command line parameters — Additional command lineparameters used for the javaw.exe command line and for the adapteritself. See specific adapter documentation for details.

• Adapter type name — (Read-only) the name of the adapter used tocreate this instance.

• Adapter version— (Read-only) the version of the adapter used to createthis instance.

• Adapter Class — (Read-only) A name that identifies the adapter class.The name depends on the type of adapter:

• Keystore Password — Optional. Data Services checks the integrity ofthe keystore data only if you enter a password. Without a password, DataServices does not check the integrity of the keystore data. This value isrequired if using HTTPS protocol to make requests.

Configuring an operation instance

To add an operation instance to an adapter instance

1. Select Adapter Instances > Job Server.2. Click the Configuration tab.3. Under Dependent Objects, click Operations.4. Click Add to configure a new operation. Or, you can click the link of an

existing operation instance to edit its configuration.5. Select an operation type from the list (Request/Reply or

Request/Acknowledge), then click Apply. The options that appeardepend on the operation specific design.

6. Complete the operation instance configuration form.7. Click Apply.

Related Topics• Configuring a Request/Reply operation instance on page 177



Configuring a Request/Reply operation instance

When creating or editing a Request/Reply operation instance, you mustcomplete the following fields:

• Operation Instance—Enter the unique operation instance name. In theDesigner, Request/Reply operation metadata object will be imported withthis name.

• Thread Count—Number of copies of Request/Reply operation to run inparallel. For parallel (asynchronous) processing of messages comingfrom a real-time service, use more than one copy. If the sequence ofmessages is important (synchronous processing), do not use more thanone thread. (Multiple copies of real-time services must be supported bymultiple copies of Request/Reply) Default is 1.

• Display Name—Enter the operation instance display name. This displayname will be visible in the Data Services Designer's metadata browsingwindow.

• Description—Enter a description of the operation instance. Thisdescription will display in the Data Services Designer's metadata browsingwindow.

• Enable—True for the Adapter SDK to start this operation instance whenadapter starts, otherwise, false.

• Target URL—URL where you want to send the HTTP request from DataServices (the HTTP server address).

• RequestMethod—The HTTP request method used to submit the request.Possible values are POST and GET.

• Content-Type—This is used to set the content type header of the request.It specifies the nature of the data by giving type and subtype identifiers.

• Content-Language—ISO code for the language in which the requestdocument is written. For example: 'en' - means that the language is Englishin one of its forms.

• Content-Encoding—Specifies the encoding mechanism used for sendingthe request. Current valid entries are x-compress and x-gzip.



• Continue If Untrusted—Specifies whether to continue the operation withan untrusted HTTP server. If True, the operation continues and if False,the operation terminates.

• Request DTD—DTD file name that defines Request XMLmessage usedin the operation.

• Request XML Root Element—Name of the XML root element in theRequest DTD.

• Reply DTD—DTD file name that defines Reply XML message used inthe operation.

• Reply XML Root Element—Name of the XML root element in the ReplyDTD.

Configuring a Request/Acknowledgement operation instance

When creating or editing a Request/Acknowledgement operation instance,you must complete the following fields:

• Operation Instance—Enter the unique operation instance name. In theDesigner, Request/Acknowledge operation metadata object is importedwith this name.

• Thread Count—Number of copies of Request/Acknowledgementoperation to run in parallel. For parallel (asynchronous) processing ofmessages coming from a real-time service, use more than one copy. Ifthe sequence of messages is important (synchronous processing), donot use more than one thread. (Multiple copies of real-time services mustbe supported by multiple copies of Request/Acknowledgement) Defaultis 1.

• Display Name—Enter the operation instance display name. This displayname will be visible in the Data Services Designer's metadata browsingwindow.

• Description—Enter a description of the operation instance. Thisdescription will display in the Data Services Designer's metadata browsingwindow.

• Enable—True for the Adapter SDK to start this operation instance whenthe adapter starts, otherwise, False.

• Target URL—URL where you want to send the HTTP request.



• Request Method—HTTP request method to be used for submitting therequest. The possible values are POST, GET.

• Content-Type—This is used to set the content type header of the request.It specifies the nature of the data, by giving type and subtype identifiers.

• Content-Language—ISO code for the language in which the requestdocument is written. For example: 'en' - means that the language is Englishin one of its forms.

• Content-Encoding—Specifies the encoding mechanism used for sendingthe request. Current valid entries are x-compress and x-gzip.

• Continue If Untrusted—Specifies whether to continue the operationinstance with an untrusted HTTP server. If True, the operation instancecontinues and if False, the operation instance terminates.

• Request DTD—DTD file name that defines Request XML message usedin this operation.

• Request XML Root Element—Name of the XML root element in theRequest DTD.

Restart the HTTP Adapter instance for the configuration changes to takeeffect.

Defining the HTTP adapter datastore

You can use the HTTP adapter in a batch or real-time data flow by selectingone of the following objects:

• An Outbound message (for Request/Acknowledge operations)

• A Message Function (for Request/Reply operations)

However before selecting these objects, you must first define an HTTPadapter datastore in the Data Services Designer, then import the operationinstances defined for the HTTP adapter instance. A data flow can then passa message to one of the adapter operation instances defined in the datastore.

To define an adapter datastore, you must:

• Define a datastore object for each adapter instance



• Define one function or one outbound message for each operation instanceto which you want to pass a message.

The following sections summarize the Designer tasks for defining an adapterdatastore. For more details, see the Data Services Designer Guide.

Related Topics• Define a datastore object on page 180

Define a datastore object

In the Designer object library, you must define a datastore object for eachadapter instance.

To define an HTTP adapter datastore1. Go to the Datastores tab in the object library, right-click and select New

from the menu.

The Create New Datastore Editor appears.

2. Name the datastore. Business Objects recommends you incorporate"HTTP" into the name.

3. For Datastore type, select Adapter.

Note: Datastore configuration options change depending on the type ofdatastore you are creating.

4. For Job Server, select the Job Server configured to handle your HTTPadapter.

5. For Adapter instance name, choose the instance name you configuredin the Administrator. For example:



6. Click OK to save values and create the datastore.

Related Topics• Configuring the HTTP adapter on page 173

Importing message functions and outbound messages to thedatastore

Data Services can pass messages from a data flow to an operation instance.You must import either a function or an outbound message (dependent onthe type of operation involved) in the Designer datastore library for eachoperation instance.

HTTP adapter operations contain the following invocation types:

Invocation TypeOperation

Message FunctionRequest/Reply

Outbound MessageRequest/Acknowledge



To import message functions and outbound messages1. In the Data Services Designer, double-click the datastore associated with

your HTTP Adapter Instance.

The Adapter Metadata Browser window opens.

2. Right-click the operation instance you want to import and select Importfrom the menu.

The operation instance you selected is added to the datastore.

You can use imported message functions and outbound messages in yourreal-time data flows.

Configuring SSL with the HTTP adapter

With Secure Socket Layer (SSL), the HTTP adapter can use secure transportover TCP/IP networks.

To connect to an SSL-enabled web application server other than thepackaged Tomcat server, you must obtain the keystore and certificate fromthat HTTP service provider. The following procedure describes how toconfigure the client.

To configure the client to access a public SSL-enabled Webserver

1. Generate the client keystore.2. Obtain a signed certificate from the SSL-enabled Web server.3. Import the certificate into the client keystore.

To use Data Services' Web server, use the following procedure.

To enable SSL on the Data Services Web server

1. Edit <JAVA_HOME>/jre/lib/security/java.security and add:security.provider.<provider no.>=com.sun.net.ssl.internal.ssl.Provider



2. At a command prompt, execute the commandkeytool -genkey -alias tomcat -keyalg RSA

In the home directory of the user under which you run it, this commandcreates a new file named *.keystore.

To specify a different location or file name, add to the keytool commandthe -keystore parameter followed by the complete path name to whereyou want to store the keystore file. You must also specify this new locationin the server-di.xml configuration file (described in the next step).

After executing the keytool command, a prompt appears requesting thekeystore password. The default password used by Tomcat is changeit(all lower case); however, you can create a custom password. You willalso need to specify the custom password in the server-di.xmlconfiguration file (described in the next step).

A second prompt requests First Name and Last Name. Enter the hostname parameter value from the Target URL property of the operationinstance.

A third prompt requests general information about this certificate includingorganization and city. This information will display to users who attemptto access a secure page in your application. Therefore, you should besure that the information provided here matches what you want users tosee.

A final prompt requests the key password, the password specificallyassociated with this certificate (as opposed to any other certificates storedin the same keystore file). You must use the same password here thatwas used for the keystore password. (The keytool prompt states that ifyou press ENTER, the password is automatically entered for you.)

If you provided the correct information, you should have created a*.keystore file with a certificate that can be used by your server.

Note: When configuring SSL on a HP-UX computer, copy the jsse.jar,jcert.jar and jnet.jar files from <LINK_DIR>/ext/lib to <JAVA_HOME>/jre/lib/ext. If the ext folder does not exist, then create itat this location, then copy the files.

3. Uncomment the following entry from the server-di.xml file. Find thesefiles in TOMCAT_HOME/conf directory.<Connector className="org.apache.tomcat.service.PoolTcpConnector">



<Parameter name="handler"value="org.apache.tomcat.service.http.HttpConnectionHandler"/><Parameter name="port" value="8443"/><Parameter name="socketFactory"value="org.apache.tomcat.net.SSLSocketFactory" /><Parameter name="keystore" value="<user.home>\.keystore" /><Parameter name="keypass" value="changeit" /></Connector>

4. Inside the Connector tag, add/update the value of the keystore andkeypass parameters. The keystore parameter should contain the .keystorefile path created in step 2. The keypass parameter should contain thepassword used to create the keystore in step 2.

5. After completing the configuration changes, restart the Data ServicesWeb server. If the Web server starts successfully, then you should beable to access any Web application supported by Tomcat via SSL.

On the client side, the HTTP Adapter client internally handles the details ofcertificate authentication by implementing the X509TrustManager interfaceand using SSLSocketFactory classes from HttpsURLConnection class.

When a HTTPS request is made to the SSL-enabled Web server, the clientwill request the server's certificate, which may be issued by some standardauthority such as VeriSign.

If the HTTP client determines that the Certificate is trusted (it checks fromits certificate store located in %LINK_DIR %\ext\Jre\lib\security), the clientreceives all the requested data from web server.

The HTTP client requires a password to query the local keystore forverification. As part of the adapter configuration, enter this password as thevalue of the keystorePassword parameter.

When encountering an untrusted certificate, the HTTP client throws theSSLException to its caller and checks the value of the continueIfUntrustedparameter. If the parameter is set to false, the SSLException displays withan informative message and gets logged in Data Services trace files. Further,the client does not receive any data from the server. If the parameter continueIfUntrusted is set to true, then Data Services logs the SSLException inerror and trace files and the client receives data from the server. Thecertificate file untrust.cer will be downloaded in the user's current workingdirectory or under LINK_DIR/bin directory.



You can import this certificate file to the JDK certificate keystore by usingthe following command:keytool -import -alias DESCRIPTION -file untrust.cer -keystore<Full path of Cacerts file in the <java.home>/lib/security/directory> cacerts -storepass changeit

where DESCRIPTION can be any text in double quotes. The storepassexpects the same password with which you created the keystore in the DataServices Web server.

You will also be prompted for a keystore password; type whatever passwordyou want. The keytool command will print out the certificate information andask you to verify it. Enter yes to complete the import process.



Troubleshooting

13

About this section

The Data Services Administrator provides status and error information. Usethis information to discover problems with your implementation and to findthe source of those problems. This sectionsection describes how to reinstallthe Web Server service and how you can use the Administrator to find andhelp resolve job processing issues.

Related Topics• Reestablishing network connections on page 188• Reinstalling the Web Server service on page 188• Finding problems on page 189• Error and trace logs on page 191• Resolving connectivity problems on page 198• Restarting the Access Server on page 200

Reestablishing network connectionsWhen you disconnect from your network and re-connect or otherwise changean IP address (Dynamic IPs), the Administrator will encounter a databaseconnection error.

To reestablish network connections for yourrepository

Either, rename the repository in the Administrator. This change forces theAdministrator to drop and recreate the connection to the database. Or, restartthe Administrator.

Reinstalling the Web Server serviceThe error "Error: 500 Location: /jsp/signin.jsp" sometimes appearswhen you try to start the Administrator even though the Data Services WebServer service is running. This error is caused by an incomplete Web Serverinstallation.


Troubleshooting13 Reestablishing network connections

To reinstall the Web Server service

1. Stop the Data Services Web Server service.2. Delete the following directories in the Data Services installation location:

• /ext/webserver/webapps/acta_web_admin

• /ext/webserver/webapps/acta_metadata_reports

3. Restart the service.

The service will re-expand the jar files and reinstall itself.

Finding problemsThe Data Services Administrator uses colored indicators to show the statusof the various system components. Generally, the indicators mean thefollowing:


13TroubleshootingFinding problems


A green indicator means the object is running properly.

A yellow indicator means some aspect of this object is notworking. Either the Access Server is in the process of its error-handling efforts to reestablish an operation, or the AccessServer is waiting for a manual intervention.

For example, when you first add a service to the AccessServer configuration, the service shows a yellow indicatoruntil you manually start the service or until you restart the Ac-cess Server.

A red indicator means one or more aspects of this object isnot working, and the error handling efforts of Access Serverwere not able to reestablish the operation of the object.

When you see a yellow or red indicator, the system requires manualintervention. You must:

• Determine which object is not operating properly

• Determine the cause of the problem

• Fix the problem

• Restart the affected service providers if necessary

To determine which object is not operating properly

1. In the Data Services Administrator, click Home.

If there is an error anywhere in the system, you will see a red indicatornext to a repository or Access Server name.

2. If you see a red indicator, click a repository or Access Server name.

The page for that repository or Access Server appears.

3. Look for another red indicator on objects listed at this level.


Troubleshooting13 Finding problems

4. If you can identify lower-level objects that have a red indicator, repeat theprevious two steps.

When you have identified the lowest level object with the error, you areready to determine the cause of the error.

To determine the cause of the error

1. Examine the error log for the affected subsystem, such as a batch job, aservice provider, an adapter interface, or for the Access Server itself.

Use the timestamp on the error entries to determine which error accountsfor the problem that you are experiencing.

2. Cross-reference the error to the trace information.

When you identify the appropriate error message, you can use thetimestamp to determine what other events occurred immediately beforethe error.

For example, if an error occurred for a specific service provider, you canuse the error timestamp in the service provider error log to look up AccessServer activities that preceded the error.

Error and trace logsThe Administrator provides access to trace and error log files for each serviceprovider, each batch job that ran, and for the Access Server. Use thesedetailed log files to evaluate and determine the cause of errors.

Batch job logs

The Batch Jobs Status page provides access to trace, monitor, and error logfiles for each batch job that ran during a specified period. For informationabout setting that period, see Setting the status interval.

For information about monitor logs, deleting logs, and the Ignore ErrorStatus button see Monitoring jobs.


13TroubleshootingError and trace logs

Related Topics• Setting the status interval on page 41• Monitoring jobs on page 91

Batch job trace logs

The trace log file lists the executed steps and the time execution began. Usethe trace log to determine where an execution failed, whether the executionsteps occurred in the order you expect, and which parts of the executionwere the most time-consuming.

To view a trace log1. Select Batch Jobs > repository.2. Under Batch Job History, find the instance of the job execution in which

you are interested.

Identify an instance by the job name, start time, etc.

3. Under Job Information for that instance, click Trace.

The Administrator opens the Job Server Trace Log Viewer.

Batch job error logs

The error log file shows the name of the object that was executing when aData Services error occurred and the text of the resulting error message. Ifthe job ran against SAP ERP or R/3 data, the error log might also includeABAP errors.

Use the error log to determine how an execution failed. If the executioncompleted without error, the error log is blank.

To view an error log1. Select Batch Jobs > repository.2. Under Batch Job History, find the instance of the job execution in which

you are interested.

Identify an instance by the job name, start time, etc.

3. Under Job Information for that instance, click Error.

The Administrator opens the Job Server Error Log Viewer.


Troubleshooting13 Error and trace logs

Service provider logs

The Service Provider Status page provides access to the error and trace logfiles for a service provider. These are the log files produced by the Job Serverthat controls the service provider.

To view the logs for a service provider

1. Select Real-Time > Access Server > Real-time Services.2. Click the name of a service.

The Administrator opens the Real-time Service Status page. This pageshows a list of service providers for the service and overall statistics forthe service.

3. Click the name of the service provider process ID in which you areinterested.

The Administrator opens the Service Provider Status page.

4. Click a link to view the desired service provider log.

To delete these logs, set the log retention period. To filter the list ofreal-time services on the Real-time Service Status page by date, set thestatus interval. .

DescriptionLink

Opens the Trace Log page for the current serviceprovider execution.

This link appears only if the real-time service is regis-tered with the Access Server.

Trace Log



DescriptionLink

Opens the Error Log page for the current serviceprovider execution.

This page lists errors generated by Data Services, bythe source or target DBMS, or the operating systemfor job execution. If the error log is empty, the job hasnot encountered errors in message processing.

This link appears only if the real-time service is regis-tered with the Access Server.

Error Log

The computer running the Job Server stores text files containing the batchand service provider trace, error and monitor logs. If you installed DataServices in the default installation location, these files are located in thefollowing directory:

LINK_DIR/Logs/JobServerName/RepoName

The name of the log file describes the contents of the file:

type_timestamp_sequence_jobname.txt

where

type : Trace, monitor, or error.

timestamp : System date and time from when the job created the log.

sequence : Number of this job related to all jobs run by this Job Serverinstance.

jobname : The name of the job instance.

Batch job trace and error logs are also available on the Log tab of theDesigner project area. To see the logs for jobs run on a particular Job Server,log in to the repository associated with the Job Server when you start DataServices Designer.



Related Topics• Setting the log retention period on page 42• Setting the status interval on page 41

Access Server logs

Trace and error logs for each Access Server are available under Real-Time> Access Server > Logs-Current and Real-Time > Access Server >Logs-History. In addition, these files are located in the Access Serverconfiguration location, which you specify when you configure the AccessServer.

Note: For remote troubleshooting, you can also connect to any AccessServer through the Administrator.

Related Topics• Adding Access Servers on page 39

To view the current day's logs

1. Select Real-time > Access Server > Logs-Current.2. This page lists the error log file followed by the trace log file.

The date of the file is included in the name:

• error_MM_DD_YYYY.log

• trace_MM_DD_YYYY.log

3. To view a file, click the name.

The Administrator shows the last 100,000 bytes of the Access Servererror log or trace log for the current date.

The error log contains error information that the Access Server generates.

The trace log contains a variety of system information. You can controlthe information the Access Server writes to the trace log.

Related Topics• To configure the trace log file on page 196



To view the previous day's logs

1. Select Real-Time > Access Server > Logs-History.2. This page lists error log files followed by trace log files. The date of the

file is included in the name:• error_MM_DD_YYYY.log

• trace_MM_DD_YYYY.log

3. To view a file, click the name.

To configure the trace log file

1. Select Real-Time > Access Server > Logs-Current.2. Click the Configuration tab.3. Under Log Contents, the Administrator lists several trace parameters

that control the information the Access Server writes to the trace file.

DescriptionName

Writes a message when an Access Server connectionto the Administrator changes.Admin

Writes a message when an Access Server exchangesinformation with a real-time service.Flow

Writes a message when an Access Server receivesrequests.Request

Writes a message when an Access Server processesauthentication information (IP addresses, user name,or password).

Security

Writes a message when an Access Server starts orstops a service.Service



DescriptionName

Writes a message when an Access Server initializes,activates, or terminates.System

4. Select the check box next to the name if you want the Access Server towrite corresponding messages to the trace log file.

5. Under Log > Tracing, select the Enabled check box.6. Click Apply.

The Administrator will change the Access Server configuration. TheAccess Server will now write the selected trace messages to the tracelog.

Note: Until you set the parameters on this page, the Access Server usesthe startup parameters to determine trace options. Each time you restart theAccess Server, the startup parameters take precedence over parametersset on this page. You can control the content of this log by setting parameterswhen configuring the Access Server.

Related Topics• Restarting the Access Server on page 200• Configuring Access Server output on page 126

To delete Access Server logs

1. Select Real-Time > Access Server > Logs-Current or Real-Time >Access Server > Logs-History.

2. Select the check box next to any log file that you want to delete.

Alternatively, to delete all the log files, select the Select all check box.

3. Click Clear or Delete.

The Administrator clears the file size for current logs and deletes theselected history files from the display and from the Access Server logdirectory.



Adapter logs

For more detailed information about an adapter or an adapter's operations,see the adapter's error and trace log files.

To view log files for an adapter instance

1. Select Adapter Instance > Job Server.2. Find the adapter instance for which you want to view logs and from the

Log Files column, click the Error Log or Trace Log link.3. The corresponding page opens.

These log files are also found in the LINK_DIR\adapters\log directory.The error log file is named adapter_instance_name_error.txt and thetrace log file is named adapter_instance_name_trace.txt.

Resolving connectivity problemsIf you have determined that you have connectivity problems among yourreal-time system components, consider the following possible failures:

• Cannot log into the Administrator

This occurs only if the local system account, that the Administrator'sinstaller uses for the Web Server, does not have privileges to launchexecutables.

Go into the Services panel and change the account used to start the DataServices Web Server service. Set properties to a user account, insteadof a local system account, and add the privilege to log in as a service.Then stop and restart the Data Services Web Server service.

• Application client cannot connect to Access Server

For example, an error appears in the logs generated by your applicationclient or in the command prompt when executing the client test utility thatlooks like this:

Error: unable to get host address


Troubleshooting13 Resolving connectivity problems

If you specified an IP address and received this error, your network mightnot support static IP address resolution. Try using the computer nameinstead.

Match the port number you specified in the Client Test utility (or in theMessage Client library call) to the Access Server's port number.

Make sure that the port you specified is not in use by other applicationson the computer where an Access Server is installed.

• Access Server cannot connect to Job Server

If this error occurs, you would see a red indicator for a service providerand an error log for the Access Server.

Match the host name and port number of the Job Server for the servicebeing called (configured in the Administrator) to the host name and portnumber that the Job Server is configured to use (as listed in the ServerManager).

Make sure that the Job Server is running by checking the Windows NTTask Manager for the Al_jobserver.exe and Al_jobservice.exeprocesses or by opening the Designer, logging in to the repository thatcorresponds to the Job Server, and looking for the Job Server icon at thebottom of the window.

• Job Server cannot start real-time service

If this error occurs, the status for the related service and its serviceprovider would be red and you would be able to open the error log filefrom the Service Provider Status page.

Make sure that the job is properly indicated for the service called in theReal-Time Service Status page.

Make sure that the real-time jobs are available in the repository associatedwith the Job Server.

Make sure the repository and the Access Server are connected to theAdministrator and that the repository is available.

If you change the password for your repository database, the job serverwill not be able to start real-time services. To fix this problem, re-registeryour repository in the Administrator and reconfigure the real-time services.

• Real-time service cannot register with Access Server


13TroubleshootingResolving connectivity problems

If this error occurs, you would see:

• A red indicator for the service provider

• An error log in the Logs-Current page (the startup timeout willeventually be triggered)

• An error log available from the Service Provider Status page.

Make sure the Access Server host name correctly identifies the computerwhere the Access Server is running.

• Access Server cannot connect back to application client

If this error occurs, you would see an error log under the Access Server'sLogs-Current node.

Make sure that the host name and port used by the message broker clientto communicate with the Access Server is correct.

Restarting the Access ServerTo restart the Access Server, you can use either of two choices:

• Controlled Restart

The Access Server responds to new and queued messages with a"shutdown" error. It waits for service providers to complete processingexisting messages, then returns the responses to those clients. Next, theAccess Server closes existing client connections (including adapters),stops, and restarts itself. Finally, the Access Server reads the currentconfiguration settings and restarts services, service providers, andadapters.

Restarting the Access Server this way requires as much time as it takesto processes requests in the system.

• Abort and Restart

The Access Server responds to new and queued messages with a"shutdown" error. It shuts down existing service providers and respondsto these messages with a "shutdown" error. Next, the Access Servercloses existing client connections (including adapters), stops, and restartsitself. Finally, the Access Server reads the current configuration settingsand restarts services, service providers, and adapters.


Troubleshooting13 Restarting the Access Server

To perform a controlled restart of the Access Server

1. Select Real-Time > Access Server > Status.2. Under Life Cycle Management, click Controlled Restart.3. Click Real-Time Services to verify that all services started properly.

The Access Server allows running services to complete and returnsincoming and queued messages to the client with a message that theAccess Server has shut down. When all services have stopped, theAccess Server stops and restarts itself. The Access Server reads the newconfiguration settings and starts services as indicated.

If all service providers started properly, the Real-Time Service Statuspage shows a green indicator next to each service name. A red indicatorsignifies that a component of the service did not start.

To perform an abort and restart of the Access Server

1. Select Real-Time > Access Server > Status.2. Under Life Cycle Management, click Abort and Restart.3. Click Real-Time Services to verify that all services started properly.


13TroubleshootingRestarting the Access Server


Troubleshooting13 Restarting the Access Server

IndexAabort

adapter instance 165RFC clients 120

Access Serveradd to the Administrator 40configuring output 126connectivity problem 198creating 127high traffic behavior 131log files, deleting 197log files, viewing 195monitoring clients 121parameters, startup 127response time 132restarting 200service request/response parameters 129,

132startup 130trace log file, configuring 196troubleshooting remotely 195

Adapter Instance node 28adapter instances

adding connections to Administrator 160error logs 198implementing, steps for 158operation instances, adding 164starting 165status of 166stopping 165trace logs 198updating configuration 161

adapter operationstatus of 166

adapter operationsadding 164starting 165statistics 166stopping 165

Administratoradding Access Servers, options 39connecting adapters 160display name 38, 144errors, finding 189, 191logging in 23navigation 25navigation tree 25password, changing 38password, defining 144Real-time node options 26roles, changing 38, 144status indicators, explanation 189Status page 25user name, changing 38user name, define for profiling 144

average processing time, service providerstatistic 135

Bbatch jobs

debugging 91, 96execution command files, for job launcher

88, 90execution options, in the Administrator 66execution status 25, 92executions, statistics for 93initiated by third-party software 82job launcher errors 90


batch jobs (continued)logs, error 97logs, trace 97scheduling 67, 83stop a running job 97

Batch node,options 26

Cclasspath 161clients

Message Broker 121, 122RFC 116, 120

communication errors, service provider statistic135

connectivity, errors 198

DData Profiler

configuring task parameters 145data generated 140defining users 143repository for 38task execution parameters 146task management parameters 148tasks status 25users 29, 38

degree of parallelism, configuring Data Profiler136

deletingbatch job log files, automatically 42batch job log files, manually 96

disablingdisplay output, Access Server 122service, automatic restart 97, 111trace log messages, adapters 161trace messages for real-time jobs 97

display name, changing in the Administrator 38display name, defining in the Administrator 144

Eenabling

service, automatic restart 97trace messages, adapters 161

error logsAccess Server, deleting 197Access Server, viewing 195adapter instances 198batch jobs, viewing 97, 192jobs, deleting automatically 42service provider 193

error replies, service provider statistic 135errors

connectivity 198finding 189, 191job launcher 90See also error logs\ 191service request/response parameters 129,

132exporting

atl file with job schedules 79

Ffile processing threads, configuring Data Profiler

136

Hhigh traffic behavior, Access Server, service

131Home 24HTTP adapter 160

IIDocs

parallel processing 119receiving from Access Server 116


Index

importingatl files, including schedules 79

inactive interval, configuring Data Profiler 136installation, testing 198invoke sleep interval, configuring Data Profiler

136

Jjob

logs, deleting automatically 42scheduling 67scheduling options 64

Job Serverconnectivity problem 198log files, viewing 193service provider, changing 109service provider, relation to 100, 103service provider, specifying 103

LLogout 24logs

monitor 94See also error logs\ 191

MManagement Console

sessions 24starting 23

max processing time, service provider statistic135

maximum concurrent tasks, configuring DataProfiler 136

Message Broker clients, examples of 121monitor log 94

Nnavigation in the Administrator 25number of days to keep results, configuring Data

Profiler 136number of distinct values, configuring Data

Profiler 136number of patterns, configuring Data Profiler

136number of records to save, configuring Data

Profiler 136

Ppassword, changing in Administrator 38password, defining in Administrator 144performance

batch job, statistics for 93service, tuning 136

ports, Job Server, service provider 103, 109processed requests, service provider statistic

135processing retry count max, service

request/response parameters 132processing timeout, service request/response

parameters 132Profiler Repositories node,options 29profiling size, configuring number of rows 136

Qqueuing timeout, service request/response

parameters 131

Rreal-time jobs

connectivity problem 198execution status 25Job Server for, changing 109log files, viewing 193


Index

real-time jobs (continued)See also service\ 100service and service provider 100service, specifying for 97supporting, tasks for 100

Real-time node,options 26recycle request count max, service

request/response parameters 130refresh interval, configuring Data Profiler 136removing services 110repository

adding to Administrator, options 33, 141viewing from the Administrator 35, 143

response time controls, service 132restarting Access Server 200RFC clients

configuration options 117IDocs, adding 119IDocs, parallel processing 119purpose of 116shutdown 120starting 119

rows per commit, configuring Data Profiler 136

Ssampling row, configuring Data Profiler 136SAP ERP and R/3, receiving IDocs from Access

Server 116scheduling jobs

creating a schedule 67deleting a schedule 78options for 64using third-party scheduler 82

server groupdefined 28determining which Job Server ran 92job launcher flag for 64using 56, 64

Server Manager, Access Server, options 126service

configuration parameters for 103

service (continued)creating 103disabling automatic restart 111IDoc, connecting to 119processing retry count max 132processing timeout 132queues for 131queuing timeout 131recycle request count max 130registration failure 130removing 110request/response, parameters for 129, 132response time controls 132service provider, adding 108service provider, relation to 100shutdown 110statistics for 133statistics, tuning 136

service provideradding to a service 108error logs 182Job Server for, specifying 103Job Server, relation to 100, 103log files, viewing 193remove or shut down 111restarting automatically 130service, relation to 100shutting down 111, 112starting 110statistics 114, 135trace logs 182

service provider[real-time jobsaaa] 100

session time out, Management Console 24Settings 24shutdown

Access Server 200adapter instance 165adapter operation 165RFC clients 120service 110service provider 111


Index

starting the Management Console 23startup timeout, service request/response

parameters 130statistics

adapter operations 166for service 133service provider 114, 135service, tuning 136

statusadapter instance 166adapter operation 166batch job executions 92client connections to an Access Server 121

status interval, setting 41stop batch jobs 97submit sleep interval, configuring Data Profiler

136

Ttesting service request/response parameters

129third party scheduler 82timeout errors, service provider statistic 135timeouts

processing 132queuing 131startup 130

trace logsAccess Server, configuring 196Access Server, deleting 197Access Server, viewing 195

trace logs (continued)adapter instances 198adapter instances, enabling messages in

161batch jobs 97batch jobs, viewing 192jobs, deleting automatically 42real-time jobs 97service provider 193

trace logs[errorsaaa] 191

trace logs[log filesaaa] 191

trace logs[troubleshootingaaa] 191

troubleshootingAccess Server, remotely 195connectivity problems 198See also error logs\ 191

UUNIX, executable batch file, for third party

scheduler 82users and roles, adding to the Administrator 38users, Administrator versus central repository

access 48

WWeb Services adapter 160


Index


Index

data services management console administrator guide3291

Documents