Fix (APAR): IZ24940 Status: Fix Product: WBI Release: 6.1.0.1 Operating System: All operating systems Supersedes Fixes: IZ21965;IZ22401;IZ22329 Pre-requisite Fixes: Exclusive-requisite Fixes: CMVC Defect: IZ24940 Byte size of APAR: 8916333 Date: 2008-07-30 Abstract: Response times of BPC API queries which include a reference to view PROCESS_INSTANCE and another view like TASK or ACTIVITY, can be improved by applying this i-Fix. Description/symptom of problem: Response times of BPC API queries which include a reference to view PROCESS_INSTANCE and another view like TASK or ACTIVITY, can be improved by applying this i-Fix. A sample query where this i-Fix applies is htm.query("TASK.TKIID", "PROCESS_INSTANCE.STATE=2", ...) or htm.query("TASK.TKIID, PROCESS_INSTANCE.PIID", null, ...) or htm.query("ACTIVITY.AIID", "PROCESS_INSTANCE.NAME='...'", ...). 1) If this i-Fix is not applied, queries as described above will take into account TASKs, ACTIVITYs etc. which can be seen by the user due to its administrative rights on process instances. 2) If this i-Fix is applied, queries as described above only take into account TASKs, ACTIVITYs etc. which can be seen by the user based on his rights on the TASKs, ACTIVITYs etc. Note that the query behavior of (2) is the query behavior as observed with WPS 6.0.2. Behavior (1) has been introduced with WPS 6.1 to make sure read authority and admin authority inherited from enclosing entities is properly reflected. Applying this i-Fix is recommended, especially when performance problems are observed. The fix reverses the behavior of BPC API queries which include process instances to the behavior as of WPS 6.0.2. Activity instance lists in BPC Explorer take into account activities which can be seen by the user due to its administrative rights on process instances. NOTE: This interim fix is applicable to both a full WPS installation as well as a WPS client installation. Directions to apply fix: For each stand-alone profile for which you configured Business Process Choreographer, do the following: 1) Make sure the application server where Business Process Choreographer is configured is stopped. 2) Install this interim fix on the stand-alone installation. - Copy pak file to the directory maintenance in the UpdateInstaller directory - Follow the Fix installation instructions that are packaged with the Fix Installer on how to install the Fix. 3) Go to the /ProcessChoreographer/config directory. 4) Depending on your platform, run one of the following commands: On Windows: ..\..\bin\wsadmin -conntype NONE -profileName -f bpeupgrade.jacl On UNIX and Linux: ../../bin/wsadmin.sh -conntype NONE -profileName -f bpeupgrade.jacl where is the name of the profile that contains the Business Process Choreographer configuration. 5) Restart the application server. For non-clustered servers in a network deployment environment 1) Stop the deployment manager. 2) Install this interim fix on the deployment manager's installation. - Copy pak file to the directory maintenance in the UpdateInstaller directory - Follow the Fix installation instructions that are packaged with the Fix Installer on how to install the Fix. 3) Start the deployment manager. This is necessary because the wsadmin scripting client must connect to the central configuration repository on the deployment manager to make the updates. 4) For each node with one or more servers where you configured Business Process Choreographer do the following: a) Make sure all servers (application servers and node agents) of the node to be upgraded are stopped. b) Install this interim fix onto the node's installation. - Copy pak file to the directory maintenance in the UpdateInstaller directory - Follow the Fix installation instructions that are packaged with the Fix Installer on how to install the Fix. c) On the node, go to the /ProcessChoreographer/config directory. d) Depending on your platform, run one of the following commands: On Windows: ..\..\bin\wsadmin -profileName -f bpeupgrade.jacl On UNIX and Linux: ../../bin/wsadmin.sh -profileName -f bpeupgrade.jacl where is the name of the node's profile. If WebSphere global security is enabled, you must also supply the "-user -password " parameters where is a user ID from the user registry which has admin authority and is the password for that user ID. 5) Restart all servers of the node. For clusters in a network deployment environment 1) Stop the deployment manager. 2) Install this interim fix on the deployment manager's installation. - Copy pak file to the directory maintenance in the UpdateInstaller directory - Follow the Fix installation instructions that are packaged with the Fix Installer on how to install the Fix. 3) Start the deployment manager. This is necessary because the wsadmin scripting client must connect to the central configuration repository on the deployment manager to make the updates. 4) For each cluster where you configured Business Process Choreographer, do the following: a) For each installation that hosts cluster members do the following: - Make sure all servers (application servers and node agents) to be upgraded are stopped. - Install this interim fix onto the node's installation. - Copy pak file to the directory maintenance in the UpdateInstaller directory - Follow the Fix installation instructions that are packaged with the Fix Installer on how to install the Fix. - Restart all servers of the node. b) On the deployment manager, go to the /ProcessChoreographer/config directory. c) Depending on your platform, run one of the following commands: On Windows: ..\..\bin\wsadmin -profileName -f bpeupgrade.jacl -cluster On UNIX and Linux: ../../bin/wsadmin.sh -profileName -f bpeupgrade.jacl -cluster where is the name of the deployment manager's profile and is the name of the cluster. If WebSphere global security is enabled, you must also supply the "-user -password " parameters where is a user ID from the user registry which has admin authority and is the password for that user ID. This will upgrade the BPC enterprise applications to the interim fix level. In case of problems, look at the /logs/bpeupgrade.log file. Directions to remove fix: NOTE: FIXES MUST BE REMOVED IN THE ORDER THEY WERE APPLIED. DO NOT REMOVE A FIX UNLESS ALL FIXES APPLIED AFTER IT HAVE FIRST BEEN REMOVED. YOU MAY REAPPLY ANY REMOVED FIX. Example: If your system has fix1, fix2, and fix3 applied in that order and fix2 is to be removed, fix3 must be removed first, then fix2 may be removed For each stand-alone profile for which you configured Business Process Choreographer, do the following: 1) Make sure the application server where Business Process Choreographer is configured is stopped. 2) Uninstall this interim fix on the stand-alone installation. - Follow the Fix instructions that are packaged with the Fix Installer on how to uninstall and reinstall the Fix. 3) Go to the /ProcessChoreographer/config directory. 4) Depending on your platform, run one of the following commands: On Windows: ..\..\bin\wsadmin -conntype NONE -profileName -f bpeupgrade.jacl On UNIX and Linux: ../../bin/wsadmin.sh -conntype NONE -profileName -f bpeupgrade.jacl where is the name of the profile that contains the Business Process Choreographer configuration. 5) Restart the application server. For non-clustered servers in a network deployment environment 1) Stop the deployment manager. 2) Uninstall this interim fix on the deployment manager's installation. - Follow the Fix instructions that are packaged with the Fix Installer on how to uninstall and reinstall the Fix. 3) Start the deployment manager. This is necessary because the wsadmin scripting client must connect to the central configuration repository on the deployment manager to make the updates. 4) For each node with one or more servers where you configured Business Process Choreographer do the following: a) Make sure all servers (application servers and node agents) of the node to be upgraded are stopped. b) Uninstall this interim fix onto the node's installation. - Follow the Fix instructions that are packaged with the Fix Installer on how to uninstall and reinstall the Fix. c) On the node, go to the /ProcessChoreographer/config directory. d) Depending on your platform, run one of the following commands: On Windows: ..\..\bin\wsadmin -profileName -f bpeupgrade.jacl On UNIX and Linux: ../../bin/wsadmin.sh -profileName -f bpeupgrade.jacl where is the name of the node's profile. If WebSphere global security is enabled, you must also supply the "-user -password " parameters where is a user ID from the user registry which has admin authority and is the password for that user ID. 5) Restart all servers of the node. For clusters in a network deployment environment 1) Stop the deployment manager. 2) Uninstall this interim fix on the deployment manager's installation. - Follow the Fix instructions that are packaged with the Fix Installer on how to uninstall and reinstall the Fix. 3) Start the deployment manager. This is necessary because the wsadmin scripting client must connect to the central configuration repository on the deployment manager to make the updates. 4) For each cluster where you configured Business Process Choreographer, do the following: a) For each installation that hosts cluster members do the following: - Make sure all servers (application servers and node agents) to be upgraded are stopped. - Uninstall this interim fix onto the node's installation. - Follow the Fix instructions that are packaged with the Fix Installer on how to uninstall and reinstall the Fix. - Restart all servers of the node. b) On the deployment manager, go to the /ProcessChoreographer/config directory. c) Depending on your platform, run one of the following commands: On Windows: ..\..\bin\wsadmin -profileName -f bpeupgrade.jacl -cluster On UNIX and Linux: ../../bin/wsadmin.sh -profileName -f bpeupgrade.jacl -cluster where is the name of the deployment manager's profile and is the name of the cluster. If WebSphere global security is enabled, you must also supply the "-user -password " parameters where is a user ID from the user registry which has admin authority and is the password for that user ID. In case of problems, look at the /logs/bpeupgrade.log file. Additional steps: If you use custom jsf clients make sure that the bpcjsfcomponents.jar contained in your custom application is replaced with the bpcjsfcomponents.jar located under ProcessChoreographer/client. The WebSphere Integration Developer uses the bpcjsfcomponents.jar located under runtimes\bi_v61_stub\bpctaglib. Therefore you need to replace the bpcjsfcomponents.jar there as well in order to use the updated bpcjsfcomponents.jar for newly generated clients.