InfoSphere - User Guide PDF
InfoSphere - User Guide PDF
InfoSphere - User Guide PDF
Version 8 Release 7
User's Guide
SC19-3450-00
Note
Before using this information and the product that it supports, read the information in Notices and trademarks on page
47.
Copyright IBM Corporation 2008, 2011.
US Government Users Restricted Rights Use, duplication or disclosure restricted by GSA ADP Schedule Contract
with IBM Corp.
Contents
Chapter 1. Overview of InfoSphere
Information Server Manager . . . . . . 1
System life cycle . . . . . . . . . . . . . 1
Architecture for development, test, and production . 2
Implementing the deployment life cycle . . . . . 2
Chapter 2. Getting started in InfoSphere
Information Server Manager . . . . . . 5
Starting the InfoSphere Information Server Manager
and adding a domain . . . . . . . . . . . 5
Working in the metadata repository . . . . . . 6
Viewing asset properties . . . . . . . . . 6
Viewing job or shared container designs . . . . 6
Searching in the InfoSphere Information Server
Manager . . . . . . . . . . . . . . . 6
Advanced searching in the InfoSphere Information
Server Manager . . . . . . . . . . . . 7
Deleting assets. . . . . . . . . . . . . 7
Tutorial: Deploying assets with the InfoSphere
Information Server Manager . . . . . . . . . 8
Introduction . . . . . . . . . . . . . 8
Module 1: Setting up the tutorial . . . . . . 9
Module 2: Creating and Deploying a package . . 10
Tutorial summary . . . . . . . . . . . 14
Tutorial: Controlling asset source with the
InfoSphere Information Server Manager . . . . . 15
Introduction . . . . . . . . . . . . . 15
Module 1: Setting up the tutorial . . . . . . 16
Module 2: Implementing source control . . . . 17
Tutorial summary . . . . . . . . . . . 20
Chapter 3. Overview of deployment . . 21
Deployment packages . . . . . . . . . . . 22
Basic deployment scenario . . . . . . . . . 22
Deployment workflow. . . . . . . . . . . 22
Deployable assets . . . . . . . . . . . . 23
Defining the deployment package . . . . . . . 23
Deleting package definitions. . . . . . . . 24
Moving package definitions . . . . . . . . 25
Copying package definitions. . . . . . . . 25
Renaming package definitions . . . . . . . 25
Setting the build and deploy paths . . . . . . 25
Building deployment packages and viewing build
histories . . . . . . . . . . . . . . . 26
Deploying packages . . . . . . . . . . . 27
Deploying packages: Single system model . . . 27
Deploying packages: Multiple system, single user
model . . . . . . . . . . . . . . . 28
Deploying packages: Multiple system, multiple
user model . . . . . . . . . . . . . 29
Chapter 4. Importing and exporting
assets . . . . . . . . . . . . . . . 31
Exporting assets . . . . . . . . . . . . . 31
Importing assets . . . . . . . . . . . . . 31
Viewing the log . . . . . . . . . . . . . 32
Chapter 5. Source control of
InfoSphere DataStage and QualityStage
assets . . . . . . . . . . . . . . . 33
Installing Rational ClearCase . . . . . . . . 34
Basic source-control scenario for jobs and associated
assets . . . . . . . . . . . . . . . . 35
Navigating assets and their annotations . . . . . 36
Configuring source control for the InfoSphere
Information Server Manager . . . . . . . . . 37
Creating source control projects. . . . . . . 37
Sharing source control projects . . . . . . . 38
Importing source control projects to your workspace 38
Providing source control for assets . . . . . . 39
Submitting assets to a source control system . . 39
Updating assets in source control . . . . . . 39
Restoring assets from the source control system 40
Product accessibility . . . . . . . . 41
Accessing product documentation. . . 43
Links to non-IBM Web sites. . . . . . 45
Notices and trademarks . . . . . . . 47
Contacting IBM . . . . . . . . . . . 51
Index . . . . . . . . . . . . . . . 53
Copyright IBM Corp. 2008, 2011 iii
iv User's Guide
Chapter 1. Overview of InfoSphere Information Server
Manager
You can use the IBM
InfoSphere
and
QualityStage
assets.
The InfoSphere Information Server Manager is a deployment tool that helps you
manage your jobs and related assets. In a typical environment, you implement
your data transformation solutions across several systems, and control the
promotion of assets between these systems. At its simplest, your environment
might contain development, test, and production systems. The deployment tool is
installed on a Windows computer, and you can connect to all of your systems from
the same Manager.
You can use the deployment tool for the following tasks:
v Viewing your InfoSphere DataStage and QualityStage assets on different
metadata repositories
v Creating packages of assets that you can build and deploy on an iterative basis
v Performing simple exports and imports of assets
v Submitting assets to a source control system
You can also manage your InfoSphere DataStage and QualityStage jobs and related
assets by using the istool commands.
System life cycle
You can use the IBM InfoSphere Information Server Manager to help you
implement a system life cycle for your IBM InfoSphere DataStage and QualityStage
applications and associated assets.
Collections of related InfoSphere DataStage and QualityStage assets, such as jobs
and shared containers, are referred to as an application. As for all software
applications, InfoSphere DataStage and QualityStage applications must be
developed and maintained under a life cycle methodology to ensure quality.
The basic life cycle methodology for InfoSphere DataStage and QualityStage
applications is concerned with the development, test, release, and maintenance of
the applications. This life cycle has a minimum of these phases:
1. Development and maintenance
2. Testing
3. Production
You can implement a system life cycle based on this model, but incorporating more
phases, for example:
1. Development and maintenance
2. Integration testing
3. Quality assurance
4. User acceptance testing
5. Production
Copyright IBM Corp. 2008, 2011 1
Architecture for development, test, and production
You can place your development, test, and production systems to implement a life
cycle for your IBM InfoSphere DataStage applications.
The best practice is to have a separate physical environment for each phase in the
life cycle. The development, test, and production systems have separate physical
environments.
You can also implement a mixed system, with the development and test systems
sharing a single physical environment, and the production system in a separate
physical environment.
Do not share the development, test, and production systems in a common physical
environment.
The physical environment for each phase in the life cycle can be a loosely coupled
cluster (massively parallel processing environment (MPP), grid, cluster) rather than
a single computer.
Each phase has one or more projects to hold the assets. Move the assets between
different phases by transferring between these projects using the source control
system.
Implementing the deployment life cycle
You can define a process to implement a life cycle for your IBM InfoSphere
DataStage applications.
About this task
The process defines the responsibilities or roles of the various personnel that are
involved in the life cycle. The description of the process does not cover any formal
paperwork that an organization might require. The description focuses on the
process of moving assets through the life cycle.
Procedure
1. Developer: Develop and maintain new assets such as jobs, custom operators,
routines, and sequences in a development project. Thoroughly unit test the
assets. Fix failing tests and make enhancements. Be sure to update the job
description, which is where the change log is kept for an asset.
2. Developer: Send the individual assets to the source code system. Include an
appropriate comment that describes the change when you submit to the
source control system. An asset is serialized as an archive file, known as an
artifact. The name of the artifact is based on the asset name.
v It provides the lowest level of granularity.
v It avoids storing the same asset more than once in the source control
system.
v It prevents any unintentional overlaying of assets when you retrieve the
assets from the source control system.
3. Developer or Administrator: Define and build deployment packages to create
a snapshot of assets with executables. You can rebuild a deployment package
at any time to incorporate the latest versions of the assets, and then deploy
the package to a target system.
2 User's Guide
4. Developer or Administrator: Submit the baseline package files to source code
system.
5. Integration Tester: Deploy the packages of controlled assets and executables
into the alpha test environment.
6. Integration Tester: Run the integration tests in the alpha environment and
evaluate the results. If the tests fail, go back to step 1. If the tests succeed,
continue to the next step.
7. Integration Tester: Label the package files for promotion to beta test. Notify
the Quality Assurance Tester that the assets are ready for beta test.
8. Quality Assurance Tester: Retrieve the package files for the beta test from the
source code system and deploy them to the beta test project.
9. Quality Assurance Tester: Run the QA tests and evaluate the results. If the
tests fail, go back to the step 1. If the tests succeed, continue to the next step.
10. Quality Assurance Tester: Label the package files as ready for production.
Notify the Production Manager that the assets are ready for release.
11. Production Manager: Ensure that a system backup is made of target
production project. The backup provides a means to recover from problems
introduced by the changed code. Retrieve the package files that are labeled for
production and deploy them into the production environment. The assets are
now deployed in the production environment.
Chapter 1. Overview of InfoSphere Information Server Manager 3
4 User's Guide
Chapter 2. Getting started in InfoSphere Information Server
Manager
Before you can move assets or submit them to the source control system, you must
connect to the metadata repository that contains the assets and locate them.
You can connect to several different metadata repositories from the same
InfoSphere Information Server Manager. You can browse the repository view to
locate the various assets that you want to work with in these repositories. You can
use search facilities to locate particular assets. You can also open a job to look at its
design (although you cannot open individual stages or change job or stage
properties).
Starting the InfoSphere Information Server Manager and adding a
domain
When you start the IBM InfoSphere Information Server Manager, your first task is
to add a domain.
About this task
You can then use the domain to connect to the Services Tier that the IBM
InfoSphere DataStage and QualityStage from which you want to deploy assets
connects to.
Procedure
1. In the Start menu, select Start > All Programs > IBM InfoSphere Information
Server > Information Server Manager.
2. In the repository tree, right-click and select Add Domain. The Add Domain
dialog opens.
3. In the Add Domain dialog, specify the domain, user, password, and whether
the password is saved.
By default, the Save Password setting is cleared. You can change the default
password setting by changing the Save Password option in the Preferences
dialog.
Changing the Save Password setting in the Preferences applies to all
subsequent invocations of the Add Domain dialog. Changing the Save
Password setting in the Add Domain dialog applies only to the specified
domain. If the Save Password setting was previously enabled, then the
password is remembered.
4. After the connection information has been established, expand the list of
available projects and package definitions on the domain and select a project.
5. If you need to reconnect to a disconnected domain, right-click on it and select
Connect. Alternatively, you can double-click the disconnected domain in the
repository tree. If a valid password has been previously saved for that domain,
InfoSphere Information Server Manager connects without prompting. If there is
no previously saved password, the Edit Domain window opens and you are
prompted to enter a password.
Copyright IBM Corp. 2008, 2011 5
Working in the metadata repository
The IBM InfoSphere Information Server Manager displays a view of the metadata
repository in a tree structure.
Each domain that you add represents a connection to a metadata repository. You
expand the domain to view a node for each of the InfoSphere Information Server
engines that share the metadata repository. You expand each engine to view the
InfoSphere DataStage and QualityStage projects that belong to that engine. Expand
the projects to view the folders that contain the assets that you want to work with.
The repository view also displays deployment packages. You can expand the
Packages to view the package definitions that were created on that domain and
stored in that metadata repository.
The repository tree is the starting point for deployment and source control tasks.
You can also work on the assets themselves from the repository tree.
Viewing asset properties
View the properties of any asset in the Repository view.
About this task
The properties give details of asset name, location, and creation and modification
dates.
Procedure
1. Right-click on the item in the Repository view.
2. Select Show Properties.
What to do next
You can also view properties from the Search Results window, Package Editor
window, Source Control view, Export Editor window, or Import Editor window.
Viewing job or shared container designs
View a job or shared container design in the Repository view.
About this task
You can view the design of a job or a shared container. You cannot make changes,
or view the properties of individual stages within the job or shared container. You
can also view designs in the Search Results window, the Package Editor window,
the Export Editor window, or the Source Control view.
Procedure
1. Double-click the job or shared container, or right-click it and select Open.
2. Use the icons on the editing toolbar at the top-right corner of the window to
zoom in, zoom out, or restore the image to its actual size.
Searching in the InfoSphere Information Server Manager
The IBM InfoSphere Information Server Manager has a search box in the
upper-right corner of the application window.
6 User's Guide
About this task
You can use the search box to perform simple searches of the metadata repository
associated with the domain that you are currently viewing. You can also search the
help topics for the InfoSphere Information Server Manager.
Procedure
1. Type the search string in the search box. You can type partial strings, for
example searching for REP might return results for asset names DEPREP.IPO,
FRPREP.ICO, and so on.
2. Select Metadata Repository or Help Search from the drop-down menu to the
right of the search box.
3. Click the search icon to perform the search and display results in the Search
window.
4. If the search returns more results than can be displayed in the Search window,
click Show more in the Search window title bar to scroll through the results.
Advanced searching in the InfoSphere Information Server
Manager
You can perform advanced searches of the metadata repository. You can specify
search criteria to perform a more specific search.
Procedure
1. Open a search window by doing one of the following steps:
v Click the search icon next to the search box in the title bar.
v Select Search from the pop-up menu on the Repository view.
2. Construct your advanced search by selecting search criteria from the
drop-down lists. Use the plus and minus buttons to add or remove
search criteria. If you do not specify a specific metadata repository to search,
the repository associated with the domain that you are currently viewing is
searched.
3. Click the search button to perform the search and display results in the Search
window.
4. If the search returns more results than can be displayed in the Search window,
click Show more in the Search window title bar to scroll through the results.
5. If you have previously performed advanced searches in this window, use the
arrow buttons in the Search window title bar to navigate through different sets
of search results.
Deleting assets
You can delete jobs from the metadata repository by using the IBM InfoSphere
Information Server Manager.
About this task
You can delete jobs from the metadata repository by using the InfoSphere
Information Server Manager. You cannot delete other types of asset because there
might be dependencies on them.
You can delete jobs from the Repository view and the Search Results window.
Chapter 2. Getting started in InfoSphere Information Server Manager 7
Procedure
1. Select the job that you want to delete.
2. Right click and select Delete.
Tutorial: Deploying assets with the InfoSphere Information Server
Manager
This tutorial teaches you the basic skills that you need to use the IBM InfoSphere
Information Server Manager to create and deploy a package.
Introduction
This tutorial walks you through the creation and deployment of a package.
The aim of the tutorial is to get you familiar with deployment in the IBM
InfoSphere Information Server Manager. The tutorial guides you through the cycle
of defining a package, building it, and deploying it. The tutorial then guides you
through building and deploying the package again after some of the assets that it
contains are changed.
The setup process involves importing some sample assets into an IBM InfoSphere
DataStage and QualityStage project, and you include these assets in your
deployment package.
Learning objectives
As you work through the tutorial, you learn how to do the following tasks:
v Connect to the metadata repository of InfoSphere Information Server and import
the tutorial data to a DataStage and QualityStage project.
v Create a package and add assets to it.
v Build the package.
v Deploy the package to a target project.
v Rebuild the package after some changes were made to the assets it contains.
v Redeploy the package to the target project.
If you have access to a second InfoSphere Information Server instance, you can do
an additional task and deploy the package to this second instance.
Time required
This tutorial takes approximately 60 minutes to finish. If you explore other
concepts that are related to this tutorial, it might take longer to complete.
Audience
New users of the InfoSphere Information Server Manager who are familiar with
InfoSphere DataStage and QualityStage.
System requirements
The tutorial requires the following hardware and software:
v InfoSphere Information Server Manager installed on a Windows platform.
8 User's Guide
v IBM InfoSphere DataStage clients installed on a Windows computer (can be the
same computer where you have the InfoSphere Information Server Manager
installed).
v Connection to a computer that hosts the services tier for InfoSphere Information
Server.
v Optionally, connection to a second computer that hosts the services tier for
another installation of InfoSphere Information Server.
Prerequisites
Complete the following tasks before starting the tutorial:
v Obtain DataStage developer privileges from the InfoSphere DataStage
administrator.
v Confirm that the InfoSphere DataStage administrator created two projects for
you to work in. They first project is named Develop, and the second project is
named Test.
v Set up the tutorial data as described in the first lesson.
Module 1: Setting up the tutorial
This module explains how to set up the data needed for this tutorial.
Learning objectives
As you work through the module, you learn how to do the following tasks:
v Connect to the metadata repository of InfoSphere Information Server and import
the tutorial data to a DataStage and QualityStage project.
Lesson 1.1: Copying tutorial data
Before you begin the tutorial, you must copy the data that you intend to use to a
folder and import it.
To set up the tutorial:
1. Open the installation media on the client computer where you installed IBM
InfoSphere DataStage.
2. Create a folder on your client computer and name it dep_exercise.
3. From the installation media, copy the \TutorialData\DataStage\
dep_exercise.isx file to the dep_exercise folder.
Lesson 1.2: Starting the InfoSphere Information Server Manager
and adding a domain
The first step is to start the IBM InfoSphere Information Server Manager and
connect to a metadata repository.
The domain is used to connect to the metadata repository holding the DataStage
and QualityStage assets that you want to package and deploy.
To start the InfoSphere Information Server Manager and connect to a metadata
repository:
1. Click Start > Programs > IBM InfoSphere Information Server > IBM
InfoSphere Information Server Manager.
2. In the Repository view of the Information Server Manager application window,
right-click and click Add Domain.
Chapter 2. Getting started in InfoSphere Information Server Manager 9
3. Specify the application server, user, password, and whether the password is
saved.
4. Click OK to connect to the domain and to the metadata repository.
The InfoSphere Information Server Manager is now ready for you to import the
tutorial data.
Lesson checkpoint:
In this lesson, you started the IInfoSphere Information Server Manager.
You learned the following tasks:
v How to start the InfoSphere Information Server Manager
v How to add a domain and use it to connect to a metadata repository
Lesson 1.3: Importing the tutorial data
Before you can start the package and deploy module, you must import the assets
to be packaged.
When you package and deploy with of the IBM InfoSphere Information Server
Manager, you define a package that can easily be updated whenever assets change.
You store the package on the computer that hosts the services tier for the
InfoSphere Information Server. You can also use a simple export mechanism to
write assets to an archive file. The archive file is stored on the Windows computer
where the InfoSphere Information Server Manager is installed. This tutorial
imports the tutorial data from an archive file that is named dep_exercise.isx.
To import the tutorial data:
1. In the Repository navigation tree, right-click the domain that contains the
project to which you want to import the assets and click Import.
2. Browse for the dep_exercise.isx file that you copied in an earlier lesson. Click
Open and view details of the assets that the archive file contains.
3. Click Import to open the Import window.
4. Select Test as the target project.
5. Click OK to import the assets to the Test project.
The InfoSphere Information Server Manager is now ready for you to start work.
Lesson checkpoint:
In this lesson, you imported the tutorial data.
You learned the following task:
v How to import assets from an archive file
Module 2: Creating and Deploying a package
This module teaches you how to create and deploy a package.
Learning objectives
As you work through the tutorial, you learn how to do the following tasks:
v Create a package and add assets to it.
v Build the package.
10 User's Guide
v Deploy the package to a target project.
v Rebuild the package after some changes were made to the assets it contains.
v Redeploy the package to the target project.
If you have access to a second InfoSphere Information Server instance, you can do
an additional task and deploy the package to this second instance.
Lesson 2.1: Creating deployment packages
You define a package that contains the assets that you want to deploy.
You create a deployment package and specify which assets you want to include in
your package. You can choose to include dependent assets in the package, to
export design assets only, job executables only, or both design assets and job
executables.
To create the deployment package:
1. Open the Package window by selecting a domain to which you are connected
in the Repository tree and doing one of the following tasks:
v Click File > New > Package from the main menu.
v Right-click, and click New > Package from the pop-up menu.
2. Select the following assets from the repository tree and drag them to the
Package window:
v dep_assets/job1
v dep_assets/job2
v dep_assets/seqjob
v dep_assets/shcont1
v dep_assets/shcont2
In this tutorial, you copy assets from a single IBM InfoSphere DataStage and
QualityStage project to add to your package. You can also add assets from
multiple projects to your package. If you have multiple InfoSphere Information
Server engines that use the services tier to which you are connected, you can
add assets from projects that are associated with different engines.
3. In the Selected Items tab, expand the view of selected folders to see the assets
that are included in the package.
4. View the dependent items for the job1 job by right-clicking on job1 in the
Package window and selecting View Dependent Items.
5. View the package creation options by clicking the Options tab on the Package
window. Select the following options:
v Include executables
v Include designs
v Include dependent items
6. Enter the following text into the Package description field: Deployment package
containing assets from the deployment tutorial
7. Save your package. Name the package dep_exercise_pack. The package is
saved as an asset in the metadata repository. You can view the package in the
packages section of the repository tree.
Lesson checkpoint:
In this lesson, you created a deployment package.
Chapter 2. Getting started in InfoSphere Information Server Manager 11
You learned the following tasks:
v How to define assets to be included in a deployment package
v How to set the options for the package and provide a description
v How to save the package to the metadata repository
Lesson 2.2: Building deployment packages
After you define a package and save it in the metadata repository, you build the
package to create a package file. You can then use the package file to deploy the
packaged assets.
When you define a package, you identify the assets that you want the package to
contain. When you build the package, copies of the assets are added to the
package file. So, the assets are included in the deployment package as they are at
the time of the build, not at the time the package was defined.
To build the deployment package:
1. Click Build in the Package window.
2. Specify the label exercise_build1 and add the comment very first build.
3. Click OK to build the package.
4. When the build is finished, click the History tab. You can see the details of the
build that you just did.
Lesson checkpoint:
In this lesson, you built a deployment package.
You learned the following tasks:
v How to build a deployment package
v How to specify a build label and add a comment to that build
Lesson 2.3: Deploying packages to a different project
You deploy the package to move the assets to another project in the same metadata
repository.
You can use the IBM InfoSphere Information Server Manager to move assets
between projects in the same metadata repository in a controlled and reproducible
manner.
To deploy the package to another project:
1. Click Deploy in the Package window.
2. Select the Test project from the Project name list.
3. Select exercise_build1 in the Build label list.
4. Click OK to deploy the package.
5. When the deploy operation is finished, click OK in the Deploy Results window.
6. Click the History tab in the Package window, and note that the Deployed on
Server field now contains the details of the project that you deployed the
package to.
Lesson checkpoint:
In this lesson, you deployed a package to a different project.
You learned the following tasks:
12 User's Guide
v How to deploy a package to a different project
v How to view deployment details in the history of the package
Lesson 2.4: Changing assets and redeploying packages
After an asset in your project changes, you can rebuild the package that contains
that asset and redeploy the package.
To build the deployment package:
1. Open an IBM InfoSphere DataStage and QualityStage Designer client and
attach to the Develop project.
2. Open the Job1 job, add an annotation, and then save the job and exit the
Designer client.
3. In the IBM InfoSphere Information Server Manager, ensure that the Package
window is showing the dep_exercise_pack package.
4. Click Build.
5. Specify the label exercise_build2 and add the comment updated build.
6. Click OK to rebuild the package.
7. When the build is finished, click the History tab. Right-click on
exercise_build2 and select Compare with Previous Build.
8. In the lower half of the Package window, compare the left pane with the right
pane. In the right pane you can see Job1 displayed in green, because Job1 is
the asset that changed between your first build and your second build.
9. Click Deploy in the Package window title bar.
10. Do the following steps:
a. Select the Test project from the Project name list.
b. Select exercise_build2 from the Build label list.
c. Select the Replace existing items option.
11. Click OK to deploy the package to the target project.
12. Refresh the Test project in the Repository view.
13. Double-click the job named Job1, and verify that it now contains the
annotation that you added to the job in the Develop project.
Lesson checkpoint:
In this lesson, you rebuilt a deployment package, where the assets it contained
changed, and deployed it to the target project again.
You learned the following tasks:
v How to rebuild a deployment package
v How to compare different builds of a package
v How to redeploy the package to the target project
Lesson 2.5: Deploying packages to a different instance of
InfoSphere Information Server
In real-world environments, your development, test, and production projects are
located on different systems. This tutorial walks you through deploying your
package to a project on a separate IBM InfoSphere Information Server instance.
You can connect to several different InfoSphere Information Server systems from a
single InfoSphere Information Server Manager. You add a domain in the
InfoSphere Information Server Manager to connect to each system. You can then
Chapter 2. Getting started in InfoSphere Information Server Manager 13
expand the domain in the repository tree and view the InfoSphere Information
Server engines and associated projects, assets, and packages.
In this tutorial, you save the package as a file to a directory on your local
Windows computer. You then deploy the assets from this file to the target
metadata repository. This technique is called the deploy-from-file method. An
alternative method transfers the package file to a defined deployment directory on
the target system and deploys the assets from there. This technique is called the
deploy-from-domain method.
To deploy your package to a different InfoSphere Information Server instance:
1. In the Package window for your package, click Send.
2. In the Save As window, select a directory on your local Windows computer to
save the package file to, and click Save.
3. Add a domain and connect to the target InfoSphere Information Server
instance.
4. Select the new domain in the repository tree, right-click and click Deploy
Package From > File.
5. Browse to the directory where you saved your package file, select the package
file, and click Open.
6. In the new Package window that opens, click Deploy and select the target
project from the Project name list and a build from the Build label list.
7. Click OK to deploy the assets to the target project.
Lesson checkpoint:
In this lesson, you deployed assets to a project on a different InfoSphere
Information Server system.
You learned the following tasks:
v How to save a deployment package to a directory on the local Windows
computer.
v How to deploy the assets in that package to a project on a different InfoSphere
Information Server instance.
Tutorial summary
From the lessons in this tutorial, you learned about deployment in the IBM
InfoSphere Information Server Manager.
Lessons learned
By completing this tutorial, you learned how to:
v Connect to the metadata repository of InfoSphere Information Server and import
the tutorial data to a DataStage and QualityStage project.
v Create a package and add assets to it.
v Build the package.
v Deploy the package to a target project.
v Rebuild the package after some changes were made to the assets it contains.
v Redeploy the package to the target project.
14 User's Guide
Tutorial: Controlling asset source with the InfoSphere Information
Server Manager
This tutorial teaches you the basic skills that you need to use the IBM InfoSphere
Information Server Manager to help you implement source control for your
InfoSphere DataStage and QualityStage assets.
Introduction
This tutorial walks you through using the IBM InfoSphere Information Server
Manager to help you implement source control for your IBM InfoSphere DataStage
and QualityStage assets.
This tutorial assumes that you use the CVS source control system. If you use a
different source control system you can adapt the tutorial. Differences between
different source control systems are in the Team menu, and your source control
system commands that are in the Team menu. It is not an aim of the tutorial to
teach you basic source control methodology.
Learning objectives
As you work through the tutorial, you learn how to do the following tasks:
v Connect to the metadata repository of InfoSphere Information Server and import
the tutorial data to an InfoSphere DataStage and QualityStage project.
v Configure your InfoSphere Information Server Manager by importing a
previously defined source control project so that the deployment tool can
connect to your source control system.
v Submit an asset to the source control system
v Resubmit an updated asset to the source control system.
v Restore your updated asset in the metadata repository with the original version
from the source control system.
Time required
This tutorial takes approximately 60 minutes to finish. If you explore other
concepts that are related to this tutorial, it might take longer to complete.
Audience
New users of IBM InfoSphere Information Server Manager who are familiar with
InfoSphere DataStage and QualityStage and with the CVS source control system.
System requirements
The tutorial requires the following hardware and software:
v InfoSphere Information Server Manager that is installed on a Windows platform.
v IBM InfoSphere DataStage and QualityStage clients on a Windows computer
(which can be the same computer where you have InfoSphere Information
Server Manager installed).
v Connection to a computer that hosts the services tier for InfoSphere Information
Server.
v Connection to a CVS source control system.
Chapter 2. Getting started in InfoSphere Information Server Manager 15
Prerequisites
Complete the following tasks before starting the tutorial:
v Obtain DataStage developer privileges from the InfoSphere DataStage
administrator.
v Confirm that the administrator created a source control project for you to work
in.
v Obtain connection details for your source control system.
v Set up the tutorial data as described in the first lesson.
Module 1: Setting up the tutorial
This module explains how to set up the data needed for this tutorial.
Learning objectives
As you work through the module, you learn how to do the following tasks:
v Connect to the metadata repository of an InfoSphere Information Server system
and import the tutorial data to an InfoSphere DataStage and QualityStage
project.
Lesson 1.1: Copying tutorial data
Before you begin the tutorial, you must copy the data to be used to a folder and
import it.
To set up the tutorial:
1. Open the installation media on the client computer where you installed IBM
InfoSphere DataStage.
2. Create a folder on your client computer and name it dep_exercise.
3. From the installation media, copy the \TutorialData\DataStage\
dep_exercise.isx file to the dep_exercise folder.
Lesson 1.2: Starting the InfoSphere Information Server Manager
and adding a domain
The first step is to start the IBM InfoSphere Information Server Manager and
connect to a metadata repository.
The domain is used to connect to the metadata repository holding the IBM
InfoSphere DataStage and QualityStage assets that you want to submit to the
source control system.
To start the InfoSphere Information ServerManager and connect to a metadata
repository:
1. Select Start > Programs > IBM InfoSphere Information Server > IBM
InfoSphere Information Server Manager.
2. In the Repository view of the InfoSphere Information Server Manager
application window, right-click, and click Add Domain. The Add Domain
window opens.
3. Specify the application server, user, password, and whether the password is
saved.
4. Click OK to connect to the domain and to the metadata repository.
The InfoSphere Information Server Manager is now ready for you to import the
tutorial data.
16 User's Guide
Lesson checkpoint:
In this lesson, you started the InfoSphere Information Server Manager.
You learned the following tasks:
v How to start the InfoSphere Information Server Manager
v How to add a domain and use it to connect to a metadata repository
Lesson 1.3: Importing the tutorial data
Before you can start the source control module, you must import the assets that
you will package.
To import the tutorial data:
1. In the Repository navigation tree, right-click the domain that contains the
project to which you want to import the assets and click Import.
2. Browse for the dep_exercise.isx file that you copied in an earlier lesson. Click
Open and view details of the assets that the archive file contains.
3. Click Import to open the Import window.
4. Select Test as the target project.
5. Click OK to import the assets to the Test project.
The InfoSphere Information Server Manager is now ready for you to start work.
Lesson checkpoint:
In this lesson, you imported the tutorial data.
You learned the following task:
v How to import assets from an archive file
Module 2: Implementing source control
This module teaches you how to use the IBM InfoSphere Information Server
Manager to implement source control for your IBM InfoSphere DataStage and
QualityStage assets.
Learning objectives
As you work through the module, you learn how to do the following tasks:
v Configure your InfoSphere Information Server Manager by importing a
previously defined source control project so that the Manager can connect to
your source control system.
v Submit an asset to the source control system
v Update the asset in InfoSphere DataStage and QualityStage and resubmit the
asset to the source control system.
v Replace your updated asset with the original version from the source control
system.
Lesson 2.1: Importing the source control project from CVS
Before you can save assets to the source control system, you must import the
source control project.
The IBM InfoSphere Information Server Manager uses a source control project to
share assets with your IBM InfoSphere DataStage projects. To configure the
Chapter 2. Getting started in InfoSphere Information Server Manager 17
InfoSphere Information Server Manager, an administrator must create the source
control project and share the project under the source code control system. Sharing
the project submits it to the source code control repository.
To import the source control project from CVS into your InfoSphere Information
Server Manager workspace:
1. Select Window > Show View > Source Control Workspace.
2. Right-click and click Import.
3. Select Projects from CVS and click Next.
4. Specify details about the CVS repository that you want to connect to and click
Next.
5. Select the Use existing repository location option, select the source control
repository location from the list (if more than one location is offered, ask your
administrator which one to select), and click Next.
6. Select the Use an existing module option and select the project to import from
the list (the project has the same name as the domain that you are connected
to), then click Next.
7. Select the Check out as a project in the workspace option.
8. Click Finish to import the project into your workspace.
Lesson checkpoint:
In this lesson, you imported a source control project into your workspace. You are
now ready to start working with the source control system.
Lesson 2.2: Submitting jobs to the source control system
This lesson shows you how to submit some example jobs to the source control
system.
When you design IBM InfoSphere DataStage jobs and move them through a
product life cycle, it is a good practice to submit designs to a source control
system, so you can control master copies and track changes. In this lesson, you
submit one of the jobs in the IBM InfoSphere Information Server Manager tutorial
data.
To submit a job:
1. Connect to the domain for the metadata repository where the example jobs are.
2. In the Repository view select the job1 job
3. Right-click the selected job, and click Send to Source Control Workspace. This
action creates an archive version of the asset, known as an artifact, and writes
that artifact to the source control project.
4. Right-click the selected job again, and click Team > Commit to commit the
artifact to the source control system. If you are not already connected to the
CVS repository, a window prompts you for authentication before you can check
in the artifact.
5. In the Commit Files window, provide a comment and click Finish.
The job1 icon is annotated to show that the job is in CVS.
Lesson checkpoint:
You submitted a job to CVS.
18 User's Guide
You learned the following:
v How to send a job to the source control workspace
v How to commit the job in CVS
v How the job icon changes to illustrate that it is in CVS
Lesson 2.3: Updating jobs in the source control system
This lesson shows you how to update the job that you previously submitted to the
source control system.
By default, executable files are sent to source control when you use the Send to
Source Control Workspace command. If you do not want to include executable
files in your source control, you must disable this option in the Source Control
section of your client preferences.
If you update an asset that was previously submitted to the source control system,
then you must update that asset in the source control system too. In this lesson
you make a minor change to job1, and then resubmit it to CVS.
To upgrade the job:
1. Open job1 in the Designer client, add an annotation to the job, and then save
the job.
2. Connect to the domain for the metadata repository where the example jobs are.
3. In the Repository view, select job1.
4. Right-click, and click Send to Source Control Workspace. This action sends an
updated archive file of the asset to the source control workspace.
5. Right-click again, and click Team > Commit to commit the artifact to the CVS
source control system. If you are not already connected to the CVS repository, a
window prompts you for authentication before you can commit the artifact.
6. In the Commit Files window, provide a comment and click Finish.
The version number next to the job icon is incremented.
Lesson checkpoint:
You have submitted a job to CVS control.
You learned the following:
v How to send a job to the source control workspace
v How to commit the job in CVS
Lesson 2.4: Restoring jobs from the source control system
This lesson shows you how to restore the job to the metadata repository that you
previously submitted to the source control system.
You might want to restore an asset from the equivalent artifact in the source
control system. For example, if your metadata repository is damaged. Alternatively
you might want to restore an asset from an earlier version in the source control
system or from a different branch in the source control system.
To restore the job from CVS:
1. Open the job1 job in the Designer client, delete the annotation that you added
in the previous lesson, and then save and close the job.
Chapter 2. Getting started in InfoSphere Information Server Manager 19
2. In the InfoSphere Information Server Manager, connect to the domain for the
metadata repository where you want to restore the job to.
3. In the Repository view, select job1.
4. Right-click the select job, and click Replace with to select whether you want to
replace the asset with the equivalent artifact or with an artifact from an
alternative branch or version. For this tutorial, select Replace with > Latest
from HEAD to restore the equivalent artifact. The Replace with action copies
the selected artifact to the source control project.
5. Right-click the job again, and click Replace from Source Control Workspace to
import the artifact from the source control workspace as an asset to the
metadata repository.
6. Open job1 in the Designer client again, and note that the annotation is restored.
If the annotation does not appear to be restored, restart the Designer client to
see that the annotation is been restored.
Lesson checkpoint:
You restored a job from CVS to the metadata repository.
You learned how you can replace an asset in your metadata repository from the
source control system.
Tutorial summary
From the lessons in this tutorial, you learned about using the IBM InfoSphere
Information Server Manager to help you implement source control for your IBM
InfoSphere DataStage and QualityStage assets.
Lessons learned
By completing this tutorial, you learned how to:
v Connect to the metadata repository of InfoSphere Information Server and import
the tutorial data to an InfoSphere DataStage and QualityStage project.
v Configure your InfoSphere Information Server Manager by importing a
previously defined source control project so that the deployment tool can
connect to your source control system.
v Submit an asset to the source control system
v Resubmit an updated asset to the source control system.
v Restore your updated asset in the metadata repository with the original version
from the source control system.
20 User's Guide
Chapter 3. Overview of deployment
Use the IBM InfoSphere Information Server Manager to deploy IBM InfoSphere
DataStage and QualityStage assets between projects on the same engine tier, or
engine tiers on different InfoSphere Information Server instances.
InfoSphere Information Server Manager supports a model of separate systems that
are used to develop, test, and run InfoSphere DataStage and QualityStage jobs. The
tool facilitates this model by providing a secure, managed, and trackable method
of moving assets between the different systems.
Deployment is an iterative process. You specify the assets that you want to deploy
by defining a package. When you are ready to deploy, you build the package to
produce a package file. If necessary, you copy the package file to the target
deployment environment. You then deploy the package to move the assets to the
target system. When the assets on the source system change, there is no need to
redefine the package; all you need to do is rebuild, recopy, and redeploy the
package.
When you define the deployment package, you can select assets from multiple
InfoSphere Information Server engines and projects which share the same metadata
repository. You can also select specific assets or folders to add all items contained
in the folder.
Package
Deployment
Information
Package
Definition
Engine
Project P
Information
Server
Information
Server
Engine 1
Project A
Project B
Project C
Project D
Engine 2
Package
File
Package
File
Copy
Build
Deploy
File System
File System
File System
Phase 1:
Define the package
Phase 2:
Build the package
Phase 3:
Deploy the package
Figure 1. Package phases in the InfoSphere Information Server Manager
Copyright IBM Corp. 2008, 2011 21
Deployment packages
To move assets between different projects or different systems, you first define a
package that contains the assets that you want to move. The package is itself
stored in the same metadata repository as the assets it contains.
After you have defined a package, you build it. The build process creates a
package file which contains the assets and which can be used to deploy the assets.
Whenever the assets change on the source system, you can rebuild the package.
You specify labels for the builds, and can supply comments to help track the
information.
The various builds of the package are maintained, so that you can revert to an
earlier version of the assets by deploying an earlier build of the package.
Basic deployment scenario
This topic provides an example deployment scenario involving IBM InfoSphere
DataStage and QualityStage assets.
In a basic deployment scenario, a set of InfoSphere DataStage and QualityStage
assets is moved from a development environment to a test or production
environment. The assets make up some or all of an application, but are not
necessarily dependent on each other. The assets that make up the set might be
developed by different users, possibly in different InfoSphere DataStage and
QualityStage projects.
The set of assets is deployed as a unit to a single target system and a single
DataStage and QualityStage project. However, the set of assets can be deployed
many times to many systems or projects as it moves through its life cycle. The
target system or systems might not allow the compilation of jobs so options are
available to package execution-ready assets.
Deployment workflow
The workflow for deploying assets by using the IBM InfoSphere Information
Server Manager might involve more than one user.
In tightly controlled environments with limited access to the target and source
systems, a developer performs definition and build steps to write the package to
the file system. A production manager (or any user with access to the required
target project and the build package) then applies the package to a project on the
target system.
Deployed assets are typically read-only in every environment except development.
If an error is discovered in a test or production environment that requires a change
to an asset, or if an improved implementation of a specific asset (such as a
InfoSphere QualityStage ruleset) is created, then the change is made in the
development environment and applied to an existing package by rebuilding the
package. The updated package can be deployed into an existing project, replacing
the corresponding assets.
There might be a requirement to revert to an earlier version of a deployment. In
this case a previous build of the package is deployed, replacing the corresponding
assets.
22 User's Guide
The following diagram shows the workflow of a created package:
Deployable assets
You can deploy most types of IBM InfoSphere DataStage and QualityStage assets
with IBM InfoSphere Information Server Manager.
You can add the following types of assets to your deployment packages:
v Data elements
v Data connections
v IMS databases
v IMS viewsets
v Jobs (mainframe, parallel, sequence, and server)
v Machine profiles
v Routines (mainframe, parallel, and server)
v Shared containers (parallel and server)
v Table definitions
v Transforms
v Custom folders (external files in the InfoSphere DataStage and QualityStage
project directory)
v QualityStage rule sets
v QualityStage match specifications
v Parallel custom stages, build stages, and wrapper stages
v Parameter sets
You can also deploy job executables, which might be required where your target
domain can only run jobs and not compile them. These executables only work if
certain conditions are met (the runtime environment must be compatible).
Defining the deployment package
You must define the deployment package before you build or deploy the package.
Finish
No Yes
Yes
No
Yes
Yes
No
No
Select prior
package build
Copy
Package
File
Define
Package
*Build
Package
Deploy
Package
Make
Changes
to Objects
Add/Remove
items from
Package
Definition
Move to
another
IS
Changes
Needed?
Deploy
Another?
Rollback
Needed?
Figure 2. Deployment package workflow
Chapter 3. Overview of deployment 23
About this task
The package definition is like a contents list for the package. You specify the assets
that you want the package to contain in the package definition, which is itself
stored as an asset in the metadata repository. You can open the package definition
at any time to add or remove assets. When you later build the package, the IBM
InfoSphere Information Server Manager assembles the assets that you specified in
the package definition into a package file. You can then deploy the assets to the
target location from the package file.
Procedure
1. Open the Package window by selecting a domain to which you are connected
in the repository tree and doing one of the following tasks:
v Select File > New > Package from the main menu.
v Select New > Package from the pop-up menu.
2. To add items to the package definition, drag the items from the repository tree
to the Selected Items tab of the Package window. Alternatively, you can
right-click on the items in the Repository view or Search window, click Copy,
then right-click in the Selected Items tab of the Deployment Package window,
and click Paste.
3. In the Selected Items tab, expand the view of selected folders to see the assets
that are included in the package.
4. To view dependent items for a single item or multiple items, right-click that
asset and click View Dependent Items.
5. To include additional files from a project file system into the package definition,
right-click in the Selected Items tab, then click Add External Folder. You can
place any files that you want to include with your package in a folder in the
project folder, and then select that folder. You could, for example, use this
feature to package some simple test data for your jobs.
6. To view the package creation options, click the Options tab on the Deployment
Package window. You can use the options to write a package description, and
include job designs, executables, or dependent items in the package.
7. When you define the package contents and options, click Save and specify a
name for the package. The package appears as an asset in the packages section
of the repository tree. You can then open the package and modify it by adding
or deleting assets.
8. You can create folders in the packages section of the repository tree to help
organize your packages. To create a folder:
a. Select the Packages node in the repository tree.
b. Right-click and click New > Folder
c. In the New Folder window, specify a name for the folder and click OK.
Results
After you have created a package definition as an asset in the metadata repository,
you can perform standard operations on the package definition, such as deleting,
moving, copying, or renaming it.
Deleting package definitions
You can delete a package definition or a package folder from the repository tree.
24 User's Guide
Procedure
1. Right-click the package definition or the folder in the repository tree.
2. Click Delete.
Moving package definitions
You can move a package definition between folders in the repository tree.
Procedure
1. Right-click the package definition in the repository tree.
2. Click Move.
3. In the Move items window, select the destination folder and click OK.
Copying package definitions
You can copy a package definition and paste it under the packages node, or in a
folder in the packages node. Use this feature as a way of replicating an existing
package definition.
Procedure
1. Right-click the package definition that you want to copy.
2. Click Copy.
3. Right-click on the packages node, or on the target folder, and click Paste.
4. Rename the package definition as required.
Renaming package definitions
You can rename a package definition in the repository tree.
Procedure
1. Right-click the package definition that you want to rename.
2. Click Rename.
3. In the Rename window, type the new name and click OK.
Setting the build and deploy paths
Set the build path to specify the location where the package file is created when a
package is successfully built. If you are deploying a package from one system to
another, set the deploy path on the target system to specify the location where you
copy the package file. In a clustering environment, set the build and deploy paths
to shared directories that are accessible from every cluster node.
Procedure
1. Connect to the domains where you want to build and deploy packages.
2. Click File > Preferences
3. Click Packages in the navigation tree.
4. Select your domain from the Domain list.
5. Enter the Build path.
6. If you are planning to deploy a package from a source system to this system,
enter the Deploy path.
Chapter 3. Overview of deployment 25
Building deployment packages and viewing build histories
Before you can deploy a package, you must build it. You can use the build history
to track the state of a package.
About this task
After you have defined a package, you can build it to produce a package that is
ready for deployment. You can rebuild existing packages and redeploy them when
the items that they contain are changed. Use the build history to track the state of
a package.
If you rebuild an existing package, you can choose to replace the entire package or
just update the items in the package that have changed since the last build.
If you choose to replace the existing package, the package is completely rebuilt and
its prior build history is deleted.
Items are pulled into the build in their state at the time of the build, not in their
state at the time the package was defined.
Procedure
1. Click Build in the Package window. The Build Package window opens, where
you specify a label and comment for the build.
2. If you want to completely replace an existing package, select the Replace
existing package option.
3. Click OK to build the package.
4. When the build is finished, click the History tab. The History view displays a
list of package builds that were either created on the domain, or deployed to
the domain.
5. To compare the contents of a build with its previous build, right-click on the
build name and select Compare with Previous Build.
6. To compare the contents of two different builds, highlight both build names,
right-click, and select Compare Selected Builds. Differences between the two
builds are shown in the bottom part of the window. The left-side view shows
the older version, and the right-side view shows the newer version of the build.
On the right-side view, items in green indicate that these new items were
added in the latest build. Items shown in blue indicate that these items were
modified.
On the left-side view, items in red indicate that these items were deleted from
the newer version of the build.
These colors settings are the default. You can customize the colors in the
Preferences menu by selecting Package, then History Comparison Colors in the
navigation view.
26 User's Guide
Note: In the status bar of the IBM InfoSphere Information Server Manager
screen is the number of changes found within the chosen builds.
Deploying packages
After you have defined and built a package that contains IBM InfoSphere
DataStage and QualityStage assets, you can deploy that package by using the
deployment tool.
You can deploy packages to different projects on the same InfoSphere DataStage
and QualityStage system, or you can deploy to different InfoSphere DataStage and
QualityStage systems.
Where you are implementing a life cycle procedure, you can use the IBM
InfoSphere Information Server Manager to deploy packages to development, test,
and production systems. In this scenario, you might require that different users
package the assets on the source system, and deploy them to the target system.
Deploying packages: Single system model
In this model, a single user creates the package and deploys it to a different project
on the same IBM InfoSphere DataStage and QualityStage system. You can deploy
any package in the Repository view to the target project if the package has been
built, and if you have the authority to import into the target project.
Figure 3. Deployment package comparison view
Chapter 3. Overview of deployment 27
Procedure
1. At the top of the Package window, click Deploy.
2. In the Deploy Package window, specify the target project and the particular
build that you want to deploy to that project. If more than one build exists, you
see a drop-down list of the builds available for deployment.
3. If required, select the Replace existing items option to overwrite items in the
target project.
4. After verifying your settings, click OK, and the package is deployed to the
target project.
When the deployment is complete, a Deploy Results window opens, displaying
the status of the deployment. In this window, you can view the deployment
details, copy the details to a clipboard, or click OK to proceed.
Deploying packages: Multiple system, single user model
In this model, a user creates the package from assets on one IBM InfoSphere
DataStage and QualityStage system, and then deploys the package to a separate
DataStage and QualityStage system.
Figure 4. Deploy Package window
28 User's Guide
About this task
This model is typically used where a life cycle is implemented between
development, test, and production systems, but access to each system is not
restricted and a single user can connect to all the domains representing these
systems.
There are two stages in this package and deploy model: one to create and build the
package on the source system, and one to deploy the package directly to the target
system.
To create and build the package on the source system:
1. Connect to the domain that represents the source DataStage and QualityStage
system.
2. Create a package from assets on the source DataStage and QualityStage system,
and build the package.
3. In the Package window, click Send in the top right corner.
4. In the Save As dialog, browse to a location to save the package file to. If
possible, browse to the deployment directory on the target server, and save the
file to that location. (A deployment directory for each domain that you added
to IBM InfoSphere Information Server Manager is specified in the Preferences
window. The deployment directory is on the computer that hosts the services
tier corresponding to that domain, which can be a Windows, UNIX, or Linux
platform.)
To deploy a package to a separate target DataStage and QualityStage system:
1. If you could not send the package file directly to the target system deploy
directory, copy the file now.
2. Connect to the domain that represents the target DataStage and QualityStage
system.
3. In the Repository view, right-click the domain representing the target system
and click Deploy Package From > Domain. The Open Package from File dialog
opens, where you select a package file to deploy. The Package window then
opens for the package that you selected.
4. Click Deploy to open the Deploy Package dialog. Select the target project and
the build that you want to deploy and click OK. The package contents are
deployed to the target project.
5. To view the package build and deployment history, click the History tab.
Deploying packages: Multiple system, multiple user model
In this model, a user creates the package from assets on one IBM InfoSphere
DataStage and QualityStage system, while another user deploys the package to a
separate DataStage and QualityStage system.
About this task
There are two stages in this package and deploy model: one to create and build the
package on the source system, and one to copy the package file to the target
system and deploy it.
To create and build the package on the source system:
1. Connect to the domain that represents the source DataStage and QualityStage
system.
Chapter 3. Overview of deployment 29
2. Create a package from assets on the source DataStage and QualityStage system,
and build the package.
3. In the Package window, click Send in the top right corner.
4. In the Save As dialog, browse to a location to save the package file to.
To deploy a package to a separate target DataStage and QualityStage system:
1. Copy the package file to a directory on the target system.
2. Connect to the domain that represents the target DataStage and QualityStage
system.
3. In the Repository view, right-click the domain representing the target system
and select Deploy Package From > File. Browse for the package file to deploy
and click Open to open selected package file in the Package window.
4. Click Deploy to open the Deploy Package dialog. Select the target project and
the build that you want to deploy and click OK. The package contents are
deployed to the target project.
5. To view the package build and deployment history, click the History tab.
30 User's Guide
Chapter 4. Importing and exporting assets
You can use the IBM InfoSphere Information Server Manager to export or import
IBM InfoSphere DataStage and QualityStage assets.
You can move assets between projects without having to perform a full
deployment. This movement is useful in cases where there is a subset of project
assets that must be populated in a secondary environment. Importing and
exporting assets in this way is like creating a .dsx file from the InfoSphere
DataStage and QualityStage Designer client.
Exporting assets
You can use the IBM InfoSphere Information Server Manager to export assets to an
archive file.
About this task
Use the InfoSphere Information Server Manager to export assets to an archive file.
The archive file is stored on the Windows computer on which the InfoSphere
Information Server Manager is running. You can then import the assets in the
archive file to any project that you can connect to by using the InfoSphere
Information Server Manager.
Procedure
1. Connect to the domain that represents the InfoSphere DataStage and
QualityStage system containing the assets you want to export.
2. Right-click the project or any assets in that project that you want to export, and
then select Export to open the Export Archive window. You can drag additional
assets into the Export Archive window.
3. If required, click the Options tab and specify that executables, designs, and
dependent items are to be included in the export.
4. When your list is complete, click Export.
5. In the Export dialog, specify the target location for the archive file.
6. Click Save.
Importing assets
You can use the IBM InfoSphere Information Server Manager to import assets to a
specific project.
About this task
You can import all the assets from an archive file to which the assets have
previously been exported.
Procedure
1. Right-click the domain that contains the project to which you want to import
the assets and click Import.
Copyright IBM Corp. 2008, 2011 31
2. Browse for a file to import in the Select file to import dialog. Click Open to
open the Import Archive window and view details of the assets that the archive
file contains.
3. Click Import to open the Import dialog.
4. Select the target project and set the option to replace any existing items.
5. Click OK to import the assets to the specified target project.
Viewing the log
View the logs from IBM InfoSphere Information Server Manager from the log view
in the InfoSphere Information Server.
The Web-based log view that tracks InfoSphere Information Server activity shows
complete manager log entries, including timestamps and any related messages.
To view the import and export logs:
1. In the InfoSphere Information Server Web console, click the Administration
tab.
2. In the Administration tab, go to Log management.
3. Click Log views. You then have the option to view import, export, and
deployment logs.
32 User's Guide
Chapter 5. Source control of InfoSphere DataStage and
QualityStage assets
You can use the IBM InfoSphere Information Server Manager deployment tool to
manage your IBM InfoSphere DataStage and QualityStage jobs and associated
assets in your source control system.
The deployment tool gives you direct access to these source control systems:
Concurrent Versions System (CVS) and IBM Rational ClearCase.
In a typical DataStage and QualityStage development environment, multiple
developers connect to the same InfoSphere Information Server system, and so to
the same metadata repository. Each developer works on their own Windows
computer that has the InfoSphere DataStage and QualityStage clients installed. If
developers want to submit the assets they develop to the source control system,
they can install the deployment tool on the same computer as these clients.
The deployment tool uses an Eclipse workspace, referred to as the source control
workspace, as a local transfer area for assets being submitted to the source control
system.
Submitting an asset from the metadata repository to your source control system is
a two-step process:
1. You send the asset to the source control workspace. The deployment tool
creates an archive file from the asset, then writes the archive file to the source
control workspace.
2. You send the artifact to the source control system. The deployment tool submits
the asset archive, known as an artifact, to the source control repository.
Restoring an asset in the metadata repository from the version in the source
control repository is a two-step process:
1. You replace the artifact in the source control workspace. The deployment tool
retrieves the asset archive, known as an artifact, from the source control
repository.
2. You replace the asset in the metadata repository. The deployment tool imports
the artifact from the source control project as an asset in the metadata
repository.
The conceptual relationship between the metadata repository, the deployment tool,
and the source control repository is illustrated in the following figure:
Copyright IBM Corp. 2008, 2011 33
Installing Rational ClearCase
Install IBM Rational
ClearCase
Windows
Printed in USA
SC19-3450-00