Using the Command-Line Interface

Let’s start the Identity Management database using the command-line interface (which most database administrators prefer). The advantage of the command-line interface methods is that they can be automated using consolidated scripts. If you are using GUI interface like Enterprise Manager, the procedure remains largely manual.

Using Oracle Enterprise Manager Cloud Control

As mentioned in an earlier section, Oracle recommends you use the Enterprise Manager Cloud Control to manage and administer the Fusion Applications environment since it provides a single interface to manage, monitor, and troubleshoot your complete Oracle infrastructure. Assuming that you already have EM installed and configured to manage the Identity Management and Fusion Applications nodes, let’s explore how you can use it to start the database tier components. Refer to Chapter 14 for how to install and configure EM Cloud Control.

In order to start the listener service, log in to Oracle Enterprise Manager Cloud Control using the SYSMAN user, click on the Targets menu, and select the Listener target type in the left panel. The screen will now look as shown in Figure 12-2.

Figure 12-2. List of configured listeners

In this screen you will see the list of configured listeners, which are IDM and FA listeners in our case. Click on the IDM listener to open the IDM listener home screen, as shown in Figure 12-3.

Figure 12-3. Starting up the listener using EM Cloud Control

Click on the Oracle Listener menu followed by Control Submenu. You can see Startup/Shutdown options here. You need to select this option even if you want to start up or shut down the service. This will lead to the Host Login details screen, as shown in Figure 12-4.

Figure 12-4. Host credential screen for listener startup/shutdown

In the Host Credential screen, you will be prompted to enter the login details of the database owner user on database host. This is mandatory in order to allow the enterprise manager agent to start the database listener service on the local host. Such credentials are not only asked for listener startup/shutdown but for all the Oracle components that we will see in the coming sections. Let me explain each of these options for providing host credentials in reverse order for better understanding.

• New: Whenever you are performing any administrative action on a component, you must provide the host login details for the first time since there are no default stored credentials. Once you select New and enter the login details, you can save the credential with a user-defined name and set as preferred credential if required.

• Named: After you save the host credential in a user defined name, you need not enter it again but you can select the Named option and then select the credential from the drop-down menu, as shown in Figure 12-4. Use this option when you have stored the login details in the user defined name but have not yet selected it as the preferred credential.

• Preferred: After a named credential is set as preferred one, you need not select new or named credentials, but just use the preferred credentials screen to go with the default value for the component and move to next screen immediately. Until the preferred credentials are set, the screen will display the message “Default preferred credentials are not set.”

Once you select appropriate credential in this screen, click Login. Confirm the action on the next screen in order to let the EM perform the required action (start the listener in this case) on the host.

Now let’s look at how to start up the IDM database using the EM Cloud Control interface. On any screen of EM, click on Targets menu followed by the Databases menu option. This will launch the Databases dashboard, as shown in Figure 12-5.

Figure 12-5. Databases common dashboard

This screen shows a consolidated summary of all the Oracle Databases configured with Enterprise Manager along with the status, availability in the last 24 hours, any critical incidents noticed in the last seven days, and so on. You may notice the list of databases configured with EM in the database status region of the screen. As you can see, the current status of the database is Down. Click on the database name idmdb to launch the home screen of the idmdb database, as shown in Figure 12-6.

Figure 12-6. IDM database startup through EM

The database home screen may display an expected error at the top, suggesting that the agent can’t establish a connection to the database. Choose the Oracle Database ➤ Control menu. You can see Startup/Shutdown as well as blackout-related options in this submenu. A blackout is required during any maintenance period to inform the Enterprise Manager Cloud Control agent to stop collecting statistics during this period. If you shut down the database using EM earlier, then a blackout would have already been there which will be removed upon startup. We will look at blackouts in detail in the next chapter where we will discuss Fusion Applications monitoring. Click on the Startup/Shutdown from the submenu to launch the Host and Database Credential screen, as shown in Figure 12-7.

Figure 12-7. Host and database credential screen for database startup/shutdown

We have seen the Host Credential screen earlier during the Listener startup where it prompted for only host OS user login details since listener startup commands are executed directly in the host shell. However in order to start the database you will need database credentials in addition to the host credential. The host credential is used to execute the commands like sqlplus in the host default shell. Database credentials are required to connect to the database instance or to bring up the database prompt in order to execute startup/shutdown commands.

You are also prompted for preferred, named, or new credentials. Regardless of whether it is host, database, or WebLogic credential, the types of credentials remain the same, as explained earlier. Create the new or use existing credential for the host and database. Make sure you select a database user with sysdba privilege for the startup/shutdown tasks.

The next screen will show the SQL being executed in the background (which is essentially STARTUP only) followed by the screen that allows you to select specific startup options like Restricted mode or STARTUP FORCE, which can abort the running instance that’s hung and start it up. (Use this with caution and only when there is no other option available.) You can also specify a custom parameter file if you want to troubleshoot some issue.

This concludes the database startup on the IDM database tier. Now let’s move on to Identity Management components startup.

Manual Startup

Although Oracle suggests you use startall.sh and stopall.sh scripts (generated during the provisioning process) in order to start or stop the Identity Management environment, we must also see how to start the components manually. We will explain the pros and cons of each methods at the end of the startup section. You may also prefer to use the script we provided in this section or create your own to start the environment.

Oracle WebLogic Server Domain for Identity Management

In an environment where Oracle Enterprise Manager Cloud Control is not installed, WebLogic Administrator Server must be started using the command line. However, the managed servers can be started using the Administrator console as well.

WebLogic Administration Server

We must start the IDM Domain Administration Server first on the node where Admin Server has been configured. Since in this case we have all IDM domain components configured on the same node we will start the same on IDM host but you may need to make sure in your environment which host is configured to run the admin server.

To start admin server, run the startWebLogic.sh script from <IDM_CONFIG>/domains/IDMDomain/. As mentioned earlier, we need to use the nohup command to let it run after leaving the session as well. You can monitor the nohup.out log file for the progress of the startup command.

[fusion@idmhost ∼]$ cd /app/oracle/config/domains/IDMDomain/bin

[fusion@idmhost ∼]$ nohup ./startWebLogic.sh &

Note that we can start up the WebLogic domain including the administration and managed servers using Enterprise Manager as well. We will see the same in the next section of starting up the WebLogic Managed Servers.

WebLogic Managed Servers

By default the following four managed servers are configured for the Identity Management domain. If you have created a high available setup then you may have more than one managed server for each WebLogic cluster. By default, the Oracle Identity Federation (wls_oif1) managed server remains down unless you have enabled and configured it.

1. Oracle Directory Service Manager (wls_ods1)

2. Oracle SOA (wls_soa1)

3. Oracle Identity Manager (wls_oim1)

4. Oracle Access Manager (wls_oam1)

We have three options to start these managed servers up manually. Of course if you use automatic startup script (startall.sh) explained later in this section, you need not use any of these manual methods

1. Using the command-line interface.

We can use the default WebLogic scripts to start up the managed servers. The prerequisite for using this method is to have the boot identity file called boot.properties present in the IDMDomain/server/<ManagedServer>/security directory to allow the script to run in the background without prompting for WebLogic server username and password. Since this file contains encrypted credentials, it is safe to keep the credentials stored in this file. If the boot identity file does not exist for any managed server then you can choose any of the following methods to create one.

a. Copy the existing boot.properties from a different directory.

You can copy the existing boot.properties file from the <ManagedServer>/data/nodemanager directory or from the AdminServer/security directory to the <ManagedServer>/security directory. This file already has encrypted credentials to start up the managed server and the same file can be copied safely.

b. Create a new boot.properties file.

If there is no existing boot.properties file, then you can simply create a new file with the following content.

username=weblogic_idm

password=<passwordof the weblogic_idm user>

Save this content as the boot.properties file. After you start the managed server the first time, it will automatically encrypt the values in the file. Now let’s start each managed server one by one. The advantage of using this method is that you need not wait for one managed server to start up completely before proceeding to the next one. Make sure to use nohup and & in order to let it run in background after you leave the session. Use the startManagedWebLogic.sh script in the <IDM_CONFIG>/domains/IDMDomain/bin directory to start the managed servers. Although Oracle’s automated startup script uses the following order to start the managed servers, you can start in any order.

[fusion@idmhost ∼]$ cd /app/oracle/config/domains/IDMDomain/bin

Let’s first start Oracle Directory service manager.

[fusion@idmhost ∼]$ nohup ./startManagedWebLogic.sh wls_ods1 &

Now start Oracle SOA.

[fusion@idmhost ∼]$ nohup ./startManagedWebLogic.sh wls_soa1 &

Now let’s start Oracle Identity Manager.

[fusion@idmhost ∼]$ nohup ./startManagedWebLogic.sh wls_oim1 &

Finally, start the Oracle Access Manager.

[fusion@idmhost ∼]$ nohup ./startManagedWebLogic.sh wls_oam1 &

You must allow some time for the managed servers to start up. You can refer to the log file nohup.out as well as the individual managed server log file at the IDMDomain/servers/<ManagedServer>/log directory to check the status until it shows the message, “Server started in RUNNING mode.”

2. Using the Oracle WebLogic administration console.

You can use Oracle WebLogic administration console to start or restart an individual managed server. But when you need to start all the managed servers then this is not the recommended method since it requires Admin Server and OAM Managed Server to be running due to the admin console’s integration with Access Manager. Additionally it requires Oracle HTTP server to be running since Oracle Access Manager SSO Login page uses HTTP server as front end. At the same time you also need to make sure that the Node Manager has been started on the hosts. However using the WebLogic admin console is very handy while manually stopping or starting individual managed servers and most administrators find this interface a simple and quick way to manage WebLogic managed servers.

Log in to the administration console at http://<AdminHost>:<AdminPort>/console as the administration user. For example, log in to http://idmhost.paramlabs.com:7777/console using the weblogic_idm user. Once you’re logged in, you will see the WebLogic Administration console homepage. Click on the Servers link in the main area or in the left panel under Domain Structure region using IDMDomain ➤ Environment ➤ Servers. You will see the summary of all WebLogic servers in the IDMDomain, as shown in Figure 12-8.

Figure 12-8. Starting up the managed servers using the administration console

As you can see, AdminServer and wls_oam1 are already started. You can start other managed servers from here by selecting them and clicking on the Start button. The next screen will prompt for confirmation to start the selected managed servers.

3. Using Oracle Enterprise Manager Cloud Control.

Once you have configured Oracle Identity Management environment with Enterprise Manager Cloud Control, you can manage each of its components using EM. Note that this includes all middleware components including HTTP Server, OID, OVD and so on, as well and not just WebLogic servers. In this example, let’s see how to start up IDM WebLogic domain using EM Cloud Control. The advantage of this method is that we need not start the administration server in advance since starting through EM provides us an option to start the complete WebLogic domain including the admin server at the same time.

Log in to Enterprise Manager Cloud Control URL at http://<em_host>:<em_port>/em using EM Administration user SYSMAN. Once you have logged on to EM, click on the Middleware link from the Targets menu. You will see Middleware summary screen, as shown in Figure 12-9.

Figure 12-9. Selecting a IDM domain target in Enterprise Manager

You will see all the configured and monitored WebLogic domains in the Details table. Click on the IDMDomain target in this list to bring up the Identity Management homepage.

The Identity Management homepage allows you start up, shut down, and manage the complete WebLogic domain, individual WebLogic managed servers, application deployments, directory services, and even web tiers. You will use this screen very often during monitoring, troubleshooting, and day to day administration of Identity Management components. This example shows you how to start the complete WebLogic domain called IDMDomain, including the administration server as well as managed servers that are enabled for startup.

In the Target Navigation region, click on the WebLogic domain followed by the IDMDomain node of the navigation tree. You can see individual managed servers under IDMDomain but we will select IDMDomain only at this moment. It will bring up the home screen for IDM WebLogic domain, as shown in Figure 12-10.

Figure 12-10. Starting up IDM domain using EM

From the IDM domain homepage, choose WebLogic Domain ➤ Control. This will bring up the Startup/Shutdown and Create/End Blackout options. Click on Startup to initiate the startup of the WebLogic domain. Alternatively you can also click the Startup button located at top of the page. This will lead to Startup options and Credentials screen, as shown in Figure 12-11.

Figure 12-11. IDM domain startup options and credentials screen

Let’s look at the important startup options required in the domain startup screen.

• End Blackout After Starting Up: By default it is checked. If you are planning to perform further maintenance even after starting up then you can uncheck this. Remember to manually end the blackout using the Control Submenu ➤ End Blackout option.

• Include Administration Server: This is very useful option since unlike the WebLogic administration console, EM allows you to start up the domain including admin server. This is also mandatory option while selecting Administration Server Process Control Method discussed next. The same option is available while shutting down as well so you can choose whether to keep admin server running after shutting down the managed servers or not.

• Process Control Method: Since Enterprise Manager effectively runs the commands at the host using EM agent, the process control method defines the set of commands to be executed in order to start the domain up. It provides the following three process control methods to choose from.

  • Administration Server: This is the default method that uses the currently running administration server to manage startup of the managed servers. You need to make sure that admin server is already running or the checkbox Include Administration Server is selected.

  • Default Script: If you want to use the default WebLogic scripts like startWeblogic.sh and startManagedWebLogic.sh to start up the domain. Make sure that the boot.properties file is available for each managed server and if the file is not available then you can create one as discussed in earlier section.

  • Custom Script: You can also create your own custom script to start up the managed servers. For example, if you want to add custom logging information or you have some pre- and post-scripts to be executed along with managed servers, you can do so here.

• Host Credentials: Similar to what we have seen in database and listener startup, here also you need to specify the IDM host credentials regardless of the process control method selected. In this example we have configured the idmhost as the IDMDomain host. It requires the credentials for this host in order to start the WebLogic server. Here also we need to select from the Preferred or Named credentials as discussed earlier. If no stored credential is available (for first time startup) then you can create a New stored named credential.

• Oracle WebLogic Domain Credentials: This frame shows up only when you have selected Administration Server as the Process Control method for starting up the managed servers. You need to specify the WebLogic administration server credentials (weblogic_idm user in this case) and save it as a Named or Preferred credential for future startup tasks.

Automated Startup

Having seen the manual methods of starting Identity Management components, let’s now look at the automated and simpler methods of starting the components. Why did we look at the manual methods first? The answer is that because in automated startup we are simply going to put the manual methods in automation scripts so it is important to understand the manual startup first followed by understanding the automated startup.

We have two options to start up the Identity Management environment automatically or by calling a single script. The first option is provided by Oracle itself while the second option is to create your own automated startup script. Why do we need to create automated startup script? Well, you will see the advantage of one method over the other soon.

Validating the Identity Management Nodes OS Processes

Once you have issued startup commands for the Identity Management components, you must make sure that the components have indeed started up. If you are starting the components manually, you could validate each component immediately after staring up but if you are starting up using single consolidated script then you can validate the OS processes one by one, as follows.

Comparison of Different Startup Options for Identity Management

At this stage you have learned about the various methods of starting up an Identity Management environment. While all of these methods are valid and applicable for starting up the components, you may want to choose one option for your routine operations. Table 12-1 shows a quick comparison of the methods with the pros and cons of each. You may want to choose one of the methods based on your priorities and organizational practice.

Using the Command-Line Interface

Once again, similar to the Identity Management database, we can comfortably start the Fusion Applications Transaction database using the command-line interface. This is the method that most database administrators prefer depending on whether the database is single node or RAC. Let’s quickly go through the steps followed by the consolidated startup scripts.

Using the Oracle Enterprise Manager Cloud Control

Assuming that you have already configured Fusion Applications database and other components with Enterprise Manager Cloud Control 12c and you have already gone through the database and middleware startup using EM in previous section, we will keep the sections related to EM-based startup compact to avoid any repetition.

In order to start the Fusion Applications database listener service, log in to Oracle Enterprise Manager Cloud Control using SYSMAN, click on the Targets menu and select Listener target type in the left panel. Select LISTENER_FUSIONDB_<DBHOST> from the list of targets. It will take you to the FUSIONDB listener home screen, as shown in Figure 12-12.

Figure 12-12. Fusion Applications Database Listener startup screen

Click on the Oracle Listener menu followed by the Control submenu. You can see the Startup/Shutdown options here. You need to select this option even if you want to start up or shut down the service. This will lead to the Host Login details screen, as we saw earlier during the IDM Database startup. The Host Credential screen prompts for the database owner and again we have option to select from Preferred, Named, or New credential. You can refer to the IDM Database startup section if you want to review the details of the types of credentials prompted in this screen.

Let’s now move on to Fusion Applications Transaction Database startup using the Enterprise Manager Cloud Control interface. Click on the Targets menu in the Global area of EM screen and then click on the Databases menu option. This will lead to the Database summary screen, which will display all configured databases in your environment. You will see that IDM database is already running and the fusiondb database is currently down. Click on fusiondb to launch Fusion Applications Database Home screen, as shown in Figure 12-13.

Figure 12-13. Fusion Applications Transaction database startup using EM

You may see an expected error on top suggesting the agent is unable to establish a connection to the database, which is expected since the database is down. Click on the Oracle Database menu, followed by the Control submenu, as shown in Figure 12-13. Click on Startup/Shutdown from the submenu to launch the Host and Database Credential screen. The Host/Database credentials screen is identical to what we have seen in IDM Database startup. Specify the database owner credentials as well as database user credentials with SYSDBA privilege.

In the next screen, you may see advanced startup options for the database, including specifying a custom parameters file, specifying different startup options for maintenance or force startup, and so on. Refer to the IDM Database startup section if you need more information about these screens. This concludes the startup of Fusion Applications transaction database and listener. Let’s move on to understanding the Fusion Applications Components startup.

Node Manager

Starting the Node Manager on each configured host is a prerequisite to WebLogic domains startup for all Fusion Applications startup methods that we are going to discuss. Each host has an instance directory for Node Manager already configured with a list of WebLogic managed servers to be monitored and controlled by the specific Node Manager. The following is the syntax to start up the Node Manager:

cd <APPLICATIONS_BASE>/instance/nodemanager/<hostname>

nohup ./ startNodeManagerWrapper.sh &

For example:

[fusion@fahost ∼]$ cd /app/oracle/instance/nodemanager/fahost

[fusion@fahost fahost]$ nohup ./startNodeManagerWrapper.sh &

You can check the output file called nohup.out for following line to make sure that the Node Manager has completed startup.

INFO: Secure socket listener started on port 5556

Oracle WebLogic Server Domains for Each Product Family

Unlike Oracle E-Business Suite R12.2 or Oracle Identity Management products, the Fusion Applications environment is comprised of dedicated WebLogic Server domains for each product family to allow dedicated resources, administration, and manageability for each product family. But this also means multiple managed servers for similar tasks for different product families. Although there is a slight overhead in starting up all these number of WebLogic domains and managed servers, eventually it helps a lot in the day-to-day operations. Depending on the number of products provisioned, your environment may have 25 to 65 managed servers. Since it may take quite some time to start them all up, using the default WebLogic scripts is not recommended since invoking startup script for all of them at same time can put a lot of load on the server CPUs. We will rely on Oracle’s specialized scripts or Enterprise Manager. Let’s look at the recommended methods of starting up the Fusion Applications WebLogic domains.

Using Enterprise Manager Cloud Control

As you now know, Enterprise Manager Cloud Control can start all the configured components in your environment including database, middleware, OPMN managed components and so on. However starting components through EM may be time consuming due to the number of screens to navigate through for multiple WebLogic domains and large number of managed servers. It is generally recommended to limit the use of EM Cloud Control based startup for specific domains or managed servers only but technically you can start up each domains manually using EM interface.

We have already seen Identity Management domain startup using EM Cloud Control interface and the Fusion Applications Domains startup is no different. We will quickly go through how to navigate to specific WebLogic domains and start up complete domain or specific WebLogic cluster manually using EM interface.

Log in to EM Cloud Control using the SYSMAN account and navigate to the Targets ➤ Fusion Applications Global menu item. You will see the Fusion Applications Summary screen, as shown in Figure 12-14.

Figure 12-14. Fusion Applications EM Summary screen

You can see the name of the Fusion Applications Instance Target in the Details region of the screen. In this example we have configured the same with FA_11192, as shown in Figure 12-14. You can expand the tree to see the list of product families domains installed in your environment. Click on the product domain which you want to manage and it will navigate to the Product Family home screen. Now you can click on the Target Domain name on the left panel (FA_11192_FinancialDomain in this example) followed by the WebLogic Domain ➤ <ProductDomain> sub-tree entry (FinancialDomain in this example) to bring up the specific domain’s home screen, as shown in Figure 12-15.

Figure 12-15. Fusion Applications product domain startup using EM

Either click on the Startup button located on top of the screen or use the WebLogic Domain ➤ Control ➤ Startup menu option to initiate startup of the domain. Similar to IDM domain startup screens, you will be prompted for host credentials as well as WebLogic domain credentials. Make sure you enter FAADMIN or equivalent Domain Administration username under WebLogic credentials and save the Named or Preferred credentials for future administration of the domain through EM.

Apart from starting up the complete product family WebLogic domain, Enterprise Manager also allows you to start up a specific WebLogic cluster, managed server, or even deployed Fusion cluster application (which may span across one or more WebLogic clusters). You can explore the Product Family EM Home screen for all such options. Figure 12-16 shows an example of starting up a specific WebLogic cluster (FinancialCommonCluster) using Enterprise Manager. As you can see, when you expand the product domain sub-tree, you will see the list of WebLogic clusters under that domain. Click on the cluster that you want to start up, which will bring up the home screen for that cluster. Click Startup or choose the WebLogic Cluster ➤ Control ➤ Startup menu option in order to start the cluster on all the nodes where it is configured to run.

Figure 12-16. Starting up the specific WebLogic server cluster using EM

Oracle HTTP Server

Fusion Applications web tier can be started at any point after the Identity Management environment has been started. However it is recommended to startup web tier (Oracle HTTP Server) after all the product domains have been started to restrict users from trying to access the application when it is not yet completely started. The HTTP server in Fusion Applications web tier is also integrated with Oracle access manager through WebGate plugin and all the applications related URLs are protected by the access manager. You can refer to the Oracle HTTP Server Virtual Host configuration files for each product family with the name FusionVirtualHost_<product_family>.conf at the following location. These files will include the Oracle HTTP Server aliases and WebLogic redirection rules. You can refer to these files for troubleshooting redirection related issues.

<APPLICATIONS_BASE>/instance/CommonDomain_webtier/config/OHS/ohs1/moduleconf/

Now let’s see how to start up the Oracle HTTP Server on the Fusion Applications web host. We have the following two options for OHS startup.

Oracle Business Intelligence Components

Oracle BI Instance can be started in a similar manner as we started OHS in the previous section. Here also we have two options to start up the Oracle BI processes with OPMN being the recommended option.

Global Order Promising (GOP): Only if Using Supply Chain Management (SCM)

If you have provisioned Oracle Fusion Supply Chain Management (SCM) products, then you may need to start the Global Order Promising (GOP) component. Note that although the fastartstop utility supports stopping the GOP component, it does not support startup of GOP as of the current release (Release 9). If you try to start up GOP using fastartstop.sh, you may see following error.

[Warning:1] [oracle.apps.startstop.cli.CLIParser: validateCommandLineArguments.793] [tid:10] Option Start not supported for GOP

The only supported method to start the GOP service is using OPMN-based startup as follows.

<APPLICATIONS_BASE>/instance/<gop_instance>/bin/opmnctl startall

For example:

[fusion@fahost bin]$ /app/oracle/instance/gop_1/bin/opmnctl startall

Oracle Social Network (OSN): Only If Configured

If you have configured the Oracle Social Network (OSN) then you can start it up as well using the fastartstop utility. Use the following command to start up OSN.

[fusion@fahost bin]$ ./fastartstop.sh -Start -componentType OSN

-appbase /app/oracle/fusionapps/applications

Informatica Identity Resolution (IIR): Only If Using CRM

If you have provisioned Oracle Fusion CRM modules, then you need to start up the Informatica Identity Resolution (IIR) component manually. Use the following commands to start up IIR on the CRM node. Note that IIR needs the DISPLAY variable to be set and the VNC server to be running on the selected port.

[fusion@fahost ∼]$ export DISPLAY=fahost:0

[fusion@fahost ∼]$ cd /app/oracle/InformaticaIR/bin/

[fusion@fahost bin]$ . ./setfusionEnv.sh

[fusion@fahost bin]$ ./liup

[fusion@fahost bin]$ ./idsup

[fusion@fahost bin]$ ./idsconc -a

Validating the Fusion Applications Nodes OS Processes

Of course you would have been making sure that each component has started up completely after startup. If you want to troubleshoot any false “component down alerts” and make sure that the OS process is indeed running, you can locate each process as follows.

Consolidated Startup/Shutdown Script Example

Having looked at all the methods of staring up the Fusion Applications environment, it is clear that starting the environment using the command-line interface is the best option since the fastartstop utility takes care of all the required dependencies within the products’ WebLogic domains. So let’s create a consolidated startup script to start all these components using a single script for quick and comfortable Fusion Applications environment startup. Feel free to customize this script if your OPMN components are hosted on different hosts. Also change the paths and passwords in the script as per your environment.

#!/bin/bash

# Script name: startFA.sh v1.0_TT

echo "Starting Node Manager Process"

cd /app/oracle/instance/nodemanager/fahost

nohup ./startNodeManagerWrapper.sh &

# Wait for Node Manager to complete the startup

sleep 30

# Start Product Domains

cd /app/oracle/fusionapps/applications/lcm/ad/bin/

echo "Starting all Fusion Applications Domains"

./fastartstop.sh -Start -all -username FAADMIN -appbase

/app/oracle/fusionapps/applications -startAdminServer true <<EOF_PASS

MyPassword

EOF_PASS

# Start OPMN based components

echo "Starting Web server"

/app/oracle/instance/CommonDomain_webtier/bin/opmnctl startall

echo "Starting BI services"

/app/oracle/instance/BIInstance/bin/opmnctl startall

echo "Starting informatica IR services"

#If idsconc exists with X display error then run vncserver :0 and xhost + as root and then rerun the command

export DISPLAY=fahost:0

# Start IIR only if CRM is provisioned or else comment these lins

cd /app/oracle/InformaticaIR/bin/

. ./setfusionEnv.sh

./liup

./idsup

./idsconc -a

echo "All components started"

Save this script as startFA.sh and add execute permission to the script so that you can run it from other wrapper scripts as well.

[fusion@fahost scripts]$ chmod +x startFA.sh

Comparison of Different Startup Options for Fusion Applications

While you may find one startup/shutdown method more suitable than other, all methods have certain advantages and disadvantages. Table 12-2 lists the major differences between these methods. You may decide on one method and implement it in your organization best practices accordingly since it is always recommended that everyone in the organization follow the same process to manage the applications services. Running multiple commands at same time in such a complex environment may cause unexpected issues.

Informatica Identity Resolution (IIR): Only if Using CRM

If you have provisioned CRM modules then you need to shut down the Informatica Identity Resolution component. Use the following commands to stop IIR;

[fusion@fahost ∼]$ cd /app/oracle/InformaticaIR/bin/

[fusion@fahost bin]$ . ./setfusionEnv.sh

[fusion@fahost bin]$ ./idsdown

[fusion@fahost bin]$ ./lidown

Oracle Social Network (OSN): Only if Configured

If you have configured Oracle Social Network (OSN), then you can stop it using the fastartstop utility as follows.

[fusion@fahost bin]$ ./fastartstop.sh -Stop -componentType OSN

-appbase /app/oracle/fusionapps/applications

Global Order Promising (GOP): Only if Using Supply Chain Management (SCM)

If you have provisioned the Oracle Fusion SCM modules, then you must stop the Global Order Promising (GOP) component. The recommended way to stop the GOP component is to use the Oracle Process Manager (OPMN) as follows.

[fusion@fahost bin]$ /app/oracle/instance/gop_1/bin/opmnctl stopall

opmnctl stopall: stopping opmn and all managed processes...

Unlike the startup of GOP, shutdown is already supported by the fastartstop utility. We can simply provide componentType as GOP as an argument to stop the component.

[fusion@fahost bin]$ ./fastartstop.sh -Stop -componentType GOP -username FAADMIN -appbase /app/oracle/fusionapps/applications

Oracle Business Intelligence Components

Unlike starting up an Oracle BI instance, the shutdown is relatively quick. The BI components can be stopped using either OPMN or the fastartstop utility but OPMN is the recommended option to shut down the Oracle BI components. Use the following command to stop it using OPMN.

[fusion@fahost bin]$ /app/oracle/instance/BIInstance/bin/opmnctl stopall

opmnctl stopall: stopping opmn and all managed processes...

If you want to use the fastartstop.sh script then you can specify BI as the -componentType argument, as shown in the following example.

[fusion@fahost bin]$ ./fastartstop.sh -Stop -componentType BI

-appbase /app/oracle/fusionapps/applications

Oracle HTTP Server

Ideally, Oracle HTTP Server should be stopped as soon as you plan to bring the complete Fusion Applications Environment down in order to prevent users from logging into the application. OPMN is the recommended method to stop the Oracle HTTP Server quickly. The following is the command to stop OHS using OPMN. However if you are planning to start it later using the fastartstop script then you can only stop OHS component and leave OPMN running.

[fusion@fahost ∼]$ /app/oracle/instance/CommonDomain_webtier/bin/opmnctl stopall

opmnctl stopall: stopping opmn and all managed processes...

Use the following command in order to stop OHS using the fastartstop utility instead.

[fusion@fahost ]$ ./fastartstop.sh -Stop -componentType OHS

-appbase /app/oracle/fusionapps/applications

Oracle WebLogic Server Domains for Each Product Family

We have already seen how to start WebLogic domains for all product families, selected product families as well as only a selected managed servers of specific WebLogic domains. The shutdown of WebLogic server domains can also be done in exactly same way by simply replacing -Start with the -Stop argument while running the fastartstop.sh script. Since we have already discussed multiple uses of the fastartstop utility to manage WebLogic Server domains, we will only look at how to stop all product domains. You can always manage selected servers by providing -domains argument as we have seen during startup. Note that this utility can stop the WebLogic domains on each Fusion Applications host where a product domain or WebLogic cluster is configured to run. Make sure that the Node Manager on each node is running before running this command.

[fusion@fahost bin]$ ./fastartstop.sh -Stop -all -username FAADMIN -fa_oracle_home

/app/oracle/fusionapps/applications

This will prompt for the WebLogic Servers Administration user specified above. If you want to leave the Administration Servers running then you can specify the -startAdminServer true argument at the end of the command. If you are going to restart the physical servers then you should not specify this argument.

Node Manager

Once all the WebLogic Server domains have been stopped, if you are planning to restart the servers then you can stop the Node Manager processes on each Fusion Applications host. If you are only going to bring up the components up after maintenance without restarting the physical servers then you can leave this running since you will need the Node Manager to be up in next startup cycle. At this moment there is no command to shut down Node Manager so we need to locate the background process for Node Manager and kill the process using the kill command. Following is the single-line command to do both these tasks.

[fusion@fahost ]$ ps -ef | grep NodeManager | awk '{print $2}' | xargs kill -9

Consolidated Script for Stopping the Fusion Applications Tier

For your convenience, we have created following consolidated script for stopping all the Fusion Applications components. Make sure you change the paths and password to actual ones, as per your environment. Feel free to modify the script to your requirements.

#!/bin/bash

# Script Name: StopFA.sh v1.0_TT

# StopOPMN Based Components

echo "Stopping informatica IR services"

cd /app/oracle/InformaticaIR/bin/

. ./setfusionEnv.sh

./idsdown

./lidown

echo "Stopping BI services"

/app/oracle/instance/BIInstance/bin/opmnctl stopall

echo "Stopping Web Server"

/app/oracle/instance/CommonDomain_webtier/bin/opmnctl stopall

# Stop WebLogic Server Domains

cd /app/oracle/fusionapps/applications/lcm/ad/bin/

echo "Stopping all Fusion Applications Domains"

./fastartstop.sh -Stop -all -username FAADMIN -appbase /app/oracle/fusionapps/applications <<EOF_PASS

MyPassword

EOF_PASS

# Stop Node Manager

echo "Killing Node Manager process"

ps -ef | grep NodeManager | awk '{print $2}' | xargs kill -9

echo "All components stopped"

Save this script as stopFA.sh and provide execute permission for this file as follows so that this file can be run from other wrapper scripts as well.

[fusion@fahost scripts]$ chmod +x stopFA.sh

Manual Shutdown Using Enterprise Manager Cloud Control

We have already seen how to start up Identity Management components using Enterprise Manager so we will skip a detailed discussion of the EM interface here. In order to shut down IDM components, follow these steps:

1. Log in to EM Cloud Control 12c with SYSMAN user.

2. On the home screen, click Targets menu ➤ Fusion Applications. This will open the Fusion Applications summary screen.

3. Expand the configured Fusion Applications instance in the targets details region. This will show the list of Fusion Applications Product Families configured with Enterprise Manager under the given Fusion instance. Click on any product family that you want to shutdown or manage. This will launch the Product Family Home screen.

4. In the Product Family Home screen, you can stop the entire WebLogic domain, specific WebLogic clusters, or even specific application deployments by selecting from the Regional area menu.

Automatic Shutdown Using the stopall.sh Script

The seeded shutdown script is already created by the Provisioning wizard at <IDM_CONFIG>/scripts. You can run the script as follows in order to shut down the Identity Management application and web tier components sequentially. Note that the password arguments are optional so if you do not specify them on the command line then the script will prompt you for them. If you are planning to run the script inside the consolidated shutdown script, you need to provide the passwords as an argument but you must check your IT security policy before specifying passwords in the plaintext script file.

<IDM_CONFIG>/scripts/stopall.sh <Node Manager Password> <WebLogic Administrator Password>

In following example, we are not specifying the passwords on the command line.

[fusion@idmhost ∼]$ cd /app/oracle/config/scripts/

[fusion@idmhost scripts]$ ./stopall.sh

Enter NodeManager Password:

Enter Weblogic Admin Password:

Manual Shutdown Using Command Line

If you want to stop all the Identity Management components manually or if you want to stop only a few selected components, then use the following manual shutdown methods.

Manual Shutdown Using Enterprise Manager Cloud Control

We have already seen how to start up Identity Management Components using Enterprise Manager so we will skip a detailed discussion of EM interface here. In order to shut down IDM components, follow these steps:

1. Log in to EM Cloud Control 12c with the SYSMAN user.

2. On the home screen, choose Targets menu ➤ Middleware. This will open the Fusion Applications Middleware targets summary screen.

3. From the Oracle Fusion Middleware Farm targets listed in the main screen, click on the IDM Middleware farm. This will open IDM Middleware targets home screen.

4. In the IDM Middleware targets screen you can stop the WebLogic domain, directory server, Oracle HTTP server and so on.

Planning for an Implementation

If you are navigating to Fusion Functional Setup work area for the first time or if you have not created an implementation project, then you will be presented with the default Getting Started with Oracle Fusion Applications screen. Once you have initiated or completed an applications implementation project, the default screen will change to the All Tasks screen. We will look at that later. For now, let’s see what the Getting Started screen has to offer.

Figure 12-19 shows an example of the Getting Started screen in Oracle Fusion Applications 11g, Release 9. Except for a small difference, in general all the regional area tasks list between 11.1.9 and earlier releases are the same. This introductory screen provides a quick overview of the implementation process flow at the top of the page. Clicking on each step of the flow leads to the appropriate work areas for those tasks.

Figure 12-19. Getting started with Fusion Functional Setup Manager

This screen also allows you to learn about the Fusion Applications’ product offerings regardless of whether they are selected for provisioning or not. As we know already, offerings are the highest level of grouping of application functionalities that can be selected for provisioning. So offerings are the initial decision point for Fusion Applications installation and implementation. Offerings include multiple business processes, which are represented as a one or more functionalities of the particular offering.

You can navigate to any of the offerings displayed in the lower panel to access documents specific to that offering. As you can see in Figure 12-20, the Documents page provides a number of documents for each of the product offering displayed. You can download the documents in PDF, HTML, or Excel formats. These documents immensely help an implementer understand the product features, dependencies, and the complete list of tasks involved in implementation process. By reviewing these documents, you can collect business requirements for the upcoming implementation project.

Figure 12-20. Offering specific documents

Table 12-3 displays the list of documents available in this screen along with their importance.

Configuring Products for Implementation

Once you have provisioned a number of product offerings and gone through the implementation steps and other documents in Offerings Documentation page, you can go ahead and select the provisioned offerings or selected functional areas of an offering for implementation. You can launch the Configure Offerings section of the Functional Setup Manager by using the link in the process flow diagram shown on the Getting Started page or from the list of tasks in the Regional area of Setup and Maintenance application.

Figure 12-21 shows an example of Configure Offerings screen with financials, Fusion accounting hub, and HCM products provisioned. This screen displays the following information for each of these offerings for the current version of Fusion Applications.

• Offering: Name of Fusion Applications product offering with optional functional areas.

• Description: Clicking on this icon will launch a small pop-up with more details of the offering.

• Provisioned: By default the field remains blank. If the product offering has been provisioned as part of the Fusion Applications provisioning process, then it will show the value as Yes.

• Enable for Implementation: By default, all checkboxes are unchecked including those offerings that are already provisioned. Select this option if you are planning to include this product offering in your upcoming implementation plan. This will enable this product specific implementation tasks to be generated.

• Implementation Status: This field tracks the status and progress of the specific offering’s implementation. Initially all values are set to Not Started. Once the implementation is complete, it will change to Implemented.

• Select Feature Choices: This option allows you to select from a list of optional feature choices available for that specific offering. Note that the feature choices may be different for individual offerings.

Figure 12-21. Configure offerings screen

Select the checkmark against each offering and functional area that you want to implement, select, or modify the optional feature choices (explained in next section) and click the Save button to freeze the selection. Figure 12-22 shows an example of Feature Choices selection for one example offering Compensation Management.

Figure 12-22. Select Feature Choices screen

The offering includes a set of business processes that need to be implemented, whereas features are optional or alternative business rules for the selected offering. There are two types of features.

• Single feature option selection: When you must select only one feature out of a list of options, then the list is provided as a set of radio buttons to restrict the selection to one only.

• Multiple features option selection: When you are allowed to select more than one options for one feature then the list is provided as a set of checkboxes that allow multiple selections.

As you can see in Figure 12-22, in the Select Feature Choices screen you may notice a button titled Lock All Selections. Once you have selected the features, you can click this button which replaces checkboxes with text or tick marks suggesting whether the feature has been selected. Once the Feature Choices are selected, you will return to the previous screen or offering selection. Click Save once you’re finished selecting offerings and feature choices for each of the offerings.

Creating an Implementation Project

Although creating and managing an implementation project is role of Application Implementation Manager, at times it is a collaborative activity between Fusion Applications Administrators and Application Implementation Managers when it comes to creating an implementation project or necessary initial setup.

Creating and managing implementation project features are one of the biggest improvements in Fusion Applications as compared to traditional ERPs, including Oracle E-Business Suite. Fusion Applications provides a completely automated and still customizable process of creating an implementation plan for your organization’s functional setup of product offerings. Based on the offerings we have selected for implementation in earlier sections, you can automatically generate a completely transparent end-to-end implementation project that can be later driven by the implementation manager and assigned to a project team based on specific tasks and individual skillsets.

In order to create a new implementation project, you need to navigate to the Manage Implementation Projects option in the list of tasks in the Regional area first. We will discuss this screen in the next section of managing an implementation project. As shown in Figure 12-23, click on the Create icon to proceed to the Create an Implementation Project screen.

Figure 12-23. Creating a new implementation project

Creating an implementation project begins with specifying a name, description, start and end dates, and so on, for the project, as shown in Figure 12-24. The code or internal name of the project will be generated automatically based on the name provided. As you will see in next section, implementation project is managed in a typical project management methodology and you may assign the project to a specific Implementation manager and assign project deadlines here. However, the deadline can be assigned or changed later as well. Click Next once you have entered this information to proceed with selecting offerings for this implementation project. Figure 12-25 shows an example of the Offerings selection page for a new implementation project.

Figure 12-24. New Implementation Project details screen

Figure 12-25. Select offerings to implement

Based on the offerings that you have selected for implementation in an earlier screen, you may include all or some of those offerings for this implementation project. You may select all or selected functional areas of an offering as well if you do not need to use some business processes in your organization.

Once you select the offerings and click Save and Open Project, it will automatically generate a list of all end-to-end functional and technical tasks for implementing the selected offerings. If there are any tasks required to be performed on other offerings that are not selected here but the current offerings have dependency on them, then those related tasks are also added in the implementation plan.

Managing an Implementation Project

Once an implementation project has been created, you can search for it from Manage Implementation Project work area. This functionality of Functional Setup Manager immensely helps functional implementers as well as system administrators manage the implementation as a full-fledged project with assigned resources, timeframe, dependency rules, task status, management, and reporting of the complete application implementation cycle. The following are a few of the salient features of implementation project management using the Fusion functional setup manager.

• A complete project management suite to create, assign, track, share and manage the end-to-end implementation project.

• Completely automated and transparent generation of all required setup tasks.

• Single and guided tasks lists for multiple offerings.

• All prerequisites and dependencies automatically are taken in account during the tasks list generation.

• Customizable tasks list to suite organization requirements to add or remove specific task to take care of exception cases.

• Eliminates the need of looking up user guides and documentation to understand the tasks involved in an implementation.

• Allows you to navigate to the assigned task directly from the tasks list.

• Graphical interface provides a summary of the implementation project at a glance.

Let’s look at one example of implementation project. Figure 12-26 shows an implementation project for the Fusion accounting hub.

Figure 12-26. Manage implementation project

As you can see, the list of implementation tasks has been generated along with individual tasks as well as the grouping of task (also called a task list) along with current status of the task and other information like number of predecessors and name of one or resources assigned the task. Note that the automatically generated tasks list is based on the offerings that you have selected for this specific project only and not all the enabled offerings.

As mentioned earlier, the generated tasks list is aware of inter-product dependencies as well as prerequisites, so it follows a specific pattern for generating the tasks list. Figure 12-27 explains how the tasks list is generated based on multiple offerings selected from more than one product families.

Figure 12-27. Generic sequence of implementation tasks

As you can see in Figure 12-27, all the common setup tasks are listed in the beginning and gradually remaining dependent tasks are listed in order. Once the common tasks for an offering, product family, or complete selection are listed, the individual offerings related tasks are generated in sequential order. Any dependency information is also updated in the implementation plan so that the implementation manager can monitor the progress of dependency tasks accordingly.

Now the Application Implementation manager needs to assign individual tasks from the generated list to functional specialists or technical administrators based on the type of task and where it needs to be performed. However, assigning each task is not mandatory but it is recommended practice in order to make sure the project management summary provides correct status information about overall progress of the implementation project. Before assigning any task you can verify the list of roles which are allowed to perform the specific task by clicking on the Authorized Roles icon next to the task name. Figure 12-28 shows an example of authorized roles for the Run User and Roles Synchronization Process task. A user with any of the listed roles can be assigned this task.

Figure 12-28. Authorized roles for the selected process

Once you have identified the roles required for the functional users, the Fusion Applications Security Administrator can create users with such privileges and then you can assign tasks to the newly created users. While assigning tasks we can also add notes to share important project information with other resources who are part of the same implementation project. We will look at user creation and role assignment in the next chapter.

Migrate and Deploy Configuration on Other Instances

While many administrators and implementers may find the automatic implementation tasks list generation as the most useful feature of the Fusion Functional Setup Manager, especially those moving from other ERPs to Fusion, a large number of implementers may find the Migration and Deployment features of Functional Setup Manager the most useful. The prime reason is the simplicity of migrating the functional setup from one environment to another. This results in a great deal of cost savings mainly due to the amount of time saved in production deployment.

Unlike traditional ERPs, Fusion Applications allows you to export complete setup data from the development or staging instance to production for initial implementation of Fusion Applications. This process is highly customizable and we can exclude or include selected setup data during the migration. This also eliminates the need of cloning complete environments when we only need to deploy setup and configuration-related data.

Once you have completed the basic implementation on a non-production environment and you want to deploy the same setup on production, you can create a new configuration package in order to export the setup data to a downloadable file. You can begin by clicking on the Manage Configuration Packages task from the regional area of Functional Setup Manager followed by clicking the Create New Configuration package icon. This will launch the basic information screen for the New Configuration package, as shown in Figure 12-29.

Figure 12-29. Create Configuration Package: Basic Information screen

From the list of existing implementation packages, select the project you want to export to the new environment. You can select only the Setup Tasks List or include Setup Data in this package depending on your requirement. The remaining details are populated automatically but you can change them if required. Click Next to proceed to the next screen where we can select objects for inclusion in the configuration package.

Figure 12-30 shows an example of the selection section for exporting setup objects. You should leave the export sequence as is unless you have a strong reason to change it. The checkbox next to each object can be selected or deselected for inclusion or exclusion from the configuration export.

Figure 12-30. Select objects to be exported in the configuration package

Once you have created the configuration package, it is ready to be exported and you can look it up under the Manage Configuration Packages work area. In order to export the Configuration package, search for the existing packages, select the one you want to export, and click the Export Setup Data button to initiate the package export.

Note that you can export the same setup information with updated data as many times as you want. Every time you run the export process, a new version of the export file is generated. You can download the latest export file in a zipped format which contains a series of XML files. This file eventually needs to be moved to a target system where it will be imported. Once the downloaded file has been copied to the target system, it will show up in the search results on the same page in the target system. You can click on Import Setup Data to initiate the setup tasks as well as the data (if exported) import. Import will insert the new data from the export file to the target. If any setup data exists in the target, it updates the existing data while leaving the data on target unchanged.

While export or import is running, you can monitor the progress using manage export and import processes task from regional area of Functional Setup Manager. Figure 12-31 shows an example of the Manage Configuration Packages screen, which shows an exported configuration file. As you can see, once the file has been exported successfully, the status changes to Locked.

Figure 12-31. Export or import configuration packages

Introduction to Maintenance Tasks

As mentioned earlier, Functional Setup Manager has two major functionalities—functional setup of and ongoing maintenance of the applications. The maintenance tasks can be accessed from the Tasks Search panel in the Regional area or from the All Tasks tab under the Maintain Implementation Projects work area. Once you have created an implementation project, the default landing page for the Functional Setup Manager changes from Getting Started page to the All Tasks page. So whenever you access the Setup and Maintenance work area from Navigator it will launch the All Tasks page first. This is appropriate since once functional setup is completed, you need to have maintenance tasks as your default screen.

Figure 12-32 shows the default tab for Maintenance Tasks, which allows you to search and execute various administrative, functional, and developer related tasks from a single interface. You will notice that these tasks are not new for you since we have already seen setup related tasks that were automatically generated as part of implementation project, while in this case you will manually search and execute the relevant maintenance task.

Figure 12-32. Search a specific administrative task

As you can see in Figure 12-32, I have searched for the very commonly used Administrative task called Run User and Roles Synchronization Process by using a partial name in the search query. Similar to the tasks in the implementation project, you can see the Authorized Roles for the selected task. The column Permitted shows whether you already have one of the Authorized roles or not by showing tick or cross mark. You may click on Go to Task in order to navigate to the task-specific page.