IBM WebSphere Portal 6013 cluster upgrade to 6015

sexevil

　　0. Download: http://www.ibm.com/support/docview.wss?rs=688&uid=swg24021241
1. For DM: Upgrade DM to 60229(60231)
  1.1 Stop DM
  1.2 Upgrade
  1.3 Start DM
2. For each node: upgrade WAS to 60229(60231)
3. For web server: stop web server, such as IHS
4. For properties file:
  4.1 wpconfig.properties:
   4.1.1 set PortalAdminPwd,WasPassword
   4.1.2 Set WpsHostPort and XmlAccessPort to the same value in the wpconfig.properties file.
      In a clustered environment, make sure they are the same value on both nodes
   4.1.3 If you want the fix pack to update the screens in the wps.ear file, add the following line to the wpconfig.properties file:
      CopyWpsEarScreens=true
   4.1.4 The WebSphere Portal Update Installer removes plain text passwords from the wpconfig*.properties files.
       To keep these passwords in the properties files, include the following line in the wpconfig.properties file: PWordDelete=false.
   
  4.2 wpconfig_dbdomain.properties:
   4.2.1 set the passwords for the database entries
   4.2.2 setting the DbUser (database name) and DbPassword (database password) parameters in the wpconfig_dbdomain.properties file.
      On the secondary node, also provide the name of the remote database server.
   4.2.3 (optional)If using a DB2 database, ensure that the DbUser in the wpconfig_dbdomain.properties file for all domains on DB2 has DBADM authority
  4.3 wpconfig_dbtype.properties
  4.4 wpconfig_sourceDB.properties
5. For Oracle Database:
  5.1 Connect to your Oracle instance as the system database administrator.
  5.2 Write down the current open_cursors value so you can reset the value after upgrading.
  5.3 Run the "alter system set open_cursors=6000;" task. (the defualt is: "alter system set open_cursors=6000 scope=spfile;")
  5.4 (optional)Depending on the number of resources and tables in your external Oracle instance, you may need to set open_cursors to a higher value.
    If, after upgrading, you receive a "maximum open cursors exceeded" error message, increase the open_cursors value and rerun the install upgrade task
6. (not sure)Ensure themes and skins have been deployed properly in accordance with the instructions in:
    http://publib.boulder.ibm.com/infocenter/wpdoc/v6r0/index.jsp?topic=/com.ibm.wp.ent.doc/wps/dgn_dpycst.html
　　7. For each nodeagent: disable automatic synchronization on all nodes and to stop all node agents
  7.1 DM -> System Administration -> Node Agents -> (click on a)nodeagent -> File Synchronization Service -> (Uncheck the) Automatic Synchronization
  7.2 Stop each nodeagent
　　Notes:The update must be performed sequentially, not in parallel on all of the server nodes in the cluster. Update the primary node first, then the secondary node and then any subsequent nodes, one at a time, in accordance with the below instructions.
8. Perform the following steps to upgrade WebSphere Portal on the primary node
  8.1 If you are using the Web server plug-in for load balancing, perform the following steps to stop traffic to the node:
   8.1.1 In the Deployment Manager administrative console, click Servers>Clusters>cluster_name>Cluster members to obtain a view of the collection of cluster members.
   8.1.2 Locate the cluster member you are upgrading and change the value in the Configured weight column to zero.
         NOTE: Record the previous value to restore the setting when the upgrade is complete. (the defual value on each member is 2)
   8.1.3 Click Apply to apply the change.
   8.1.4 If automatic plug-in generation and propagation is disabled, manually generate and/or propagate the plugin-cfg.xml file to the Web servers.
  8.2 (optional)If you have a local Web Server on the node you are upgrading, stop the local Web server
  8.3 (optional)If necessary, perform the following steps to upgrade the WebSphere Application Server on the node:
   8.3.1 Run the following command from the was_profile_root/bin directory to stop the node agent;
      NOTE: If security is not enabled, exclude the -user and -password parameters from the command:
      Unix: ./stopNode.sh -user wpsbind -password wpsbind
   8.3.2 Upgrade WebSphere Application Server on the node, including the required interim fixes for WebSphere Portal.
   8.3.3 The wmm.jar is only updated on the WebSphere Portal server (in the AppServer/lib directory on that node).
      If the Deployment Manager uses separate AppServer binaries or is on a different server altogether,
      you must manually copy these files to the Deployment Manager's file system to update its WMM software:
       wp.wire.jar, wmm.jar, and the wmm.ejb.jar
   8.3.4 Run the following command from the was_profile_root/bin directory to start the node agent
  8.4 (optional)Customers using a production database can clear out the WebSphere Portal database tranlogs as discussed in the technote
     http://www-01.ibm.com/support/docview.wss?rs=688&uid=swg21288663
  8.5 Verify that the deployment manager and node agent for the primary node are running. If they are stopped, start them
  8.6 If using the Universal PUI, (which does not include the bundled Java environment),
        run the following command . ./setupCmdLine.sh for Unix from the was_profile_root/bin directory
        to set up the Java environment for the graphical user interface installation program.
  8.7 (command line for 8.6, optional)
     Unix: portal_server_root/update> ./updatePortal.sh -install -installDir "/opt/WebSphere/PortalServer" -fixpack -fixpackDir "/opt/WebSphere/PortalServer/update/fixpacks" -fixpackID WP_PTF_6015
  8.8 Use the following steps to remap WebSphere Portal to the external web server using the Deployment Manager Administrative Console:
     8.8.1 Navigate to Applications > Enterprise Applications.
     8.8.2 Click on the name: wps EA name.
     8.8.3 Click on Map modules to servers.
     8.8.4 Select WebSphere Portal Server (wps.war) module.
     8.8.5 From the Clusters and Servers drop-down menu, select the Web server and Portal Cluster.
     8.8.6 Click Apply.
  8.9 Regenerate the Web server plug-in. Copy the plugin-cfg.xml file to the Plugin directory. Restart the Web server.
  8.10 After the fix pack is installed, check the status of the node you are upgrading in the Deployment Manager administrative console.
     8.10.1 If the status is Not Synchronized, ensure that the node agent is running on the node and then perform the following steps:
     8.10.2 In the Deployment Manager administrative console, click System Administration>Nodes.
     8.10.3 For the node with a status of Not Synchronized, click Synchronize.
     8.10.4 After the synchronization is complete, wait at least 30 minutes before performing the next step because the ear expander is still running.
     8.10.5 Restart the WebSphere_Portal server on the primary node.
  8.11 If you stopped a local Web server, restart it now. Start the external Web server, if appropriate
  8.12 Run the following task to activate the portlets:
     Unix: ./WPSconfig.sh finalize-portlets-fixpack -DPortalAdminPwd=wpsbind
  8.13 Verify that your system is operational by entering the server's URL in a browser and logging in to browse the content
  8.14 If you are using the Web server plug-in for load balancing, perform the following steps to restore traffic to the upgraded node:
     8.14.1 In the Deployment Manager administrative console, click Servers>Clusters>cluster_name>Cluster members to obtain a view of the collection of cluster members.
     8.14.2 Locate the cluster member you upgraded and change the value in the Configured weight column back to the original value. (value=2)
     8.14.3 Click Apply to apply the change.
     8.14.4 If automatic plug-in generation and propagation is disabled, manually generate and/or propagate the plugin-cfg.xml file to the Web servers.
  8.15 (optional) open the wpconfig.properties file and include the following line in the file: PWordDelete=true
     run ./WPSconfig.sh action-delete-passwords-fixpack
  8.16 If you upgraded from a prior release with additional interim fixes installed,
      you will need to reinstall any interim fixes that were not integrated into the version 6.0.1.5 installation.
      Open the CheckAdditionalEfixes_date_time.log file, located in the portal_server_root/version/log directory,
      for the list of interim fixes that need to be reinstalled.
      Before reinstalling the interim fix, go to the
        http://www.ibm.com/software/genservers/portal/support/
      to see if there is a newer version of the interim fix because these are often specific to a version and release of the product;
      and search on the APAR number to find more information
  8.17 Enter the appropriate command to remove the work directory that was created during the upgrade
      ./WPSconfig.sh cleanup-work-dir
9. Perform the following steps to upgrade WebSphere Portal on each secondary node:
 9.1 If you are using the Web server plug-in for load balancing, perform the following steps to stop traffic to the node:
  9.1.1 In the Deployment Manager administrative console, click Servers>Clusters>cluster_name>Cluster members to obtain a view of the collection of cluster members.
  9.1.2 Locate the cluster member you are upgrading and change the value in the Configured weight column to zero.
     NOTE: Record the previous value to restore the setting when the upgrade is complete. (2)
  9.1.3 Click Apply to apply the change.
  9.1.4 If automatic plug-in generation and propagation is disabled, manually generate and/or propagate the plugin-cfg.xml file to the Web servers
  9.2 (optional)If you have a local Web Server on the node you are upgrading, stop the local Web server
  9.3 (optional)If necessary, perform the following steps to upgrade the WebSphere Application Server on the node:
   9.3.1 Run the following command from the was_profile_root/bin directory to stop the node agent;
      NOTE: If security is not enabled, exclude the -user and -password parameters from the command:
      Unix: ./stopNode.sh -user wpsbind -password wpsbind
   9.3.2 Upgrade WebSphere Application Server on the node, including the required interim fixes for WebSphere Portal.
   9.3.3 The wmm.jar is only updated on the WebSphere Portal server (in the AppServer/lib directory on that node).
      If the Deployment Manager uses separate AppServer binaries or is on a different server altogether,
      you must manually copy these files to the Deployment Manager's file system to update its WMM software:
       wp.wire.jar, wmm.jar, and the wmm.ejb.jar
   9.3.4 Run the following command from the was_profile_root/bin directory to start the node agent
  9.4 (optional)Customers using a production database can clear out the WebSphere Portal database tranlogs as discussed in the technote
     http://www-01.ibm.com/support/docview.wss?rs=688&uid=swg21288663  
  9.5 Verify that the deployment manager and node agent for the primary node are running. If they are stopped, start them
  9.6 If using the Universal PUI, (which does not include the bundled Java environment),
        run the following command . ./setupCmdLine.sh for Unix from the was_profile_root/bin directory
        to set up the Java environment for the graphical user interface installation program.
  9.7 (command line for 8.6, optional)
     Unix: portal_server_root/update> ./updatePortal.sh -install -installDir "/opt/WebSphere/PortalServer" -fixpack -fixpackDir "/opt/WebSphere/PortalServer/update/fixpacks" -fixpackID WP_PTF_6015
  9.8 Use the following steps to remap WebSphere Portal to the external web server using the Deployment Manager Administrative Console:
     9.8.1 Navigate to Applications > Enterprise Applications.
     9.8.2 Click on the name: wps EA name.
     9.8.3 Click on Map modules to servers.
     9.8.4 Select WebSphere Portal Server (wps.war) module.
     9.8.5 From the Clusters and Servers drop-down menu, select the Web server and Portal Cluster.
     9.8.6 Click Apply.
  9.9 Regenerate the Web server plug-in. Copy the plugin-cfg.xml file to the Plugin directory. Restart the Web server.
  9.10 After the fix pack is installed, check the status of the node you are upgrading in the Deployment Manager administrative console.
     9.10.1 If the status is Not Synchronized, ensure that the node agent is running on the node and then perform the following steps:
     9.10.2 In the Deployment Manager administrative console, click System Administration>Nodes.
     9.10.3 For the node with a status of Not Synchronized, click Synchronize.
     9.10.4 After the synchronization is complete, wait at least 30 minutes before performing the next step because the ear expander is still running.
     9.10.5 Restart the WebSphere_Portal server on the primary node.
  9.11 If you stopped a local Web server, restart it now. Start the external Web server, if appropriate
  9.12 Run the following task to activate the portlets:
     Unix: ./WPSconfig.sh finalize-portlets-fixpack -DPortalAdminPwd=wpsbind
  9.13 Verify that your system is operational by entering the server's URL in a browser and logging in to browse the content
  9.14 If you are using the Web server plug-in for load balancing, perform the following steps to restore traffic to the upgraded node:
     9.14.1 In the Deployment Manager administrative console, click Servers>Clusters>cluster_name>Cluster members to obtain a view of the collection of cluster members.
     9.14.2 Locate the cluster member you upgraded and change the value in the Configured weight column back to the original value. (value=2)
     9.14.3 Click Apply to apply the change.
     9.14.4 If automatic plug-in generation and propagation is disabled, manually generate and/or propagate the plugin-cfg.xml file to the Web servers.
  9.15 (optional) open the wpconfig.properties file and include the following line in the file: PWordDelete=true
     run ./WPSconfig.sh action-delete-passwords-fixpack
  9.16 If you upgraded from a prior release with additional interim fixes installed,
      you will need to reinstall any interim fixes that were not integrated into the version 6.0.1.5 installation.
      Open the CheckAdditionalEfixes_date_time.log file, located in the portal_server_root/version/log directory,
      for the list of interim fixes that need to be reinstalled.
      Before reinstalling the interim fix, go to the
        http://www.ibm.com/software/genservers/portal/support/
      to see if there is a newer version of the interim fix because these are often specific to a version and release of the product;
      and search on the APAR number to find more information
  9.17 Enter the appropriate command to remove the work directory that was created during the upgrade
      ./WPSconfig.sh cleanup-work-dir
10. Perform the following post-cluster installation upgrade steps
  10.1 (not sure,optional)You do not have to manually merge the files if you added the CopyWpsEarScreens=true line to the wpconfig.properties file.
  10.2 Perform the following steps to change the open_cursors value back to its original value if you are using an external Oracle database:
     Connect to your Oracle instance as the system database administrator.
     Run the "alter system set open_cursors=original_value" task.
     ----Shut down and then restart your Oracle instance.
  10.3 (not sure)If you are using XMLAccess remotely you will need to replace your current files wp.xml.client.jar and wp.base.jar
  10.4 (not sure)The WebSphere Portal 6.0.1.5 fix pack does not update any of the business portlets
  10.5 Perform the following steps to re-enable automatic synchronization on all nodes:
    10.5.1 In the Deployment Manager administrative console, click System Administration>Node Agents.
    10.5.2 Click nodeagent on the required node.
    10.5.3 Click File Synchronization Service on the Configuration tab.
    10.5.4 Check the Automatic Synchronization check box to enable the automatic synchronization feature and then click OK.
    10.5.5 Run the following command from the was_profile_root/bin directory to stop the node agent;
      NOTE: If security is not enabled, exclude the -user and -password parameters from the command:
      Unix: ./stopNode.sh -user wpsbind -password wpsbind
    10.5.6 Run the following command from the was_profile_root/bin directory to start the node agent:
      Unix: ./startNode.sh
    10.5.7 Repeat these steps for all remaining nodes.