Modifying individual configuration properties
Use the appropriate 100Custom.xml files in your topology to modify the configuration properties of IBM Process Server or IBM Process Center.
- Configuring transaction timeouts
In the 100Custom.xml files for Process Server and Process Center, you can add or modify configuration settings that are used to specify transaction timeouts. - Setting the save mode in IBM Process Designer
You can set the option that determines how changes in the editor are saved in the web Process Designer by adding the collaboration-mode configuration setting to the 100Custom.xml files. - Enabling browser caching to improve web Process Designer performance
In IBM BPM, there are two settings that control whether browser caching is enabled for resources in the snapshots of toolkits. When browser caching is enabled, it reduces the number of network calls for resources that are needed by editors. This improves the scalability of the editor and enhances performance. To override the default value for these settings, you can use the updateBPMConfig command to add or modify the settings in the 100Custom.xml file. - Selecting the EPV data to use in snapshot deployment or instance migration
In earlier releases of IBM BPM, the latest modified EPV data was used in snapshot deployment or instance migration. Interim fix JR47706 introduced a new behavior where the default EPV data for a target snapshot version overwrites the EPV data set in the older snapshot release. However, the later interim fix JR52960 restores the behavior where the latest modified EPV data is used rather than the default EPV data. A configurable property epv-deploy-default has been introduced to let you toggle between these two behaviors. - Deleting process instance tokens when a referenced task does not exist
In earlier releases of IBM BPM, you could not delete process instance tokens if the task that the process refers to did not exist or the process did not contain tasks. However, a configurable property force-token-action has been introduced to let you delete these orphaned process instance tokens that have no associated tasks. - Migrating a large number of inflight process instances using the Process Admin Console
When you use the Process Admin Console to migrate a large number of inflight process instances, the web browser might crash or hang because of memory issues. The server returns very large blocks of instance data to the client browser. However, a configurable property migrate-instances-count has been introduced to let you to specify the number of instances that are migrated during a single migration using the Process Admin Console. - Disabling team synchronization during process instance migration
When you migrate process instances into a new IBM Business Process Manager (BPM) snapshot on an IBM Process Server environment, groups and members are automatically reinstated that were intentionally deleted from teams to prevent them from being able to launch a business process definition (BPD). However, you can use the configurable property disable-team-sync to disable team synchronization during process instance migration and prevent deleted groups and members from being automatically added back to teams. - Monitoring and repairing user memberships for ad hoc groups
Generally, you should never change the user membership of an ad hoc group. Regardless, there are several ways to change the membership and inadvertently render the ad hoc group incorrect. Since all ad hoc groups are shared, any change in the membership means that all subsequently created tasks (which are assigned to the same list of users) are associated with the incorrect ad hoc group and its membership. However, you can monitor ad hoc groups and repair incorrect user memberships by adding the enable-ad-hoc-group-monitoring configuration setting to the 100Custom.xml files. - Controlling authenticated user access to internal service types
In earlier releases of IBM BPM, you could invoke a service by using the executeServiceByName URL and there was no access restriction based on the service type. Instead, services that were only meant for internal use were available to all authenticated users. Although this release of IBM BPM includes behavior that now validates the service type for invocations that are performed using the executeServiceByName URL, an enforce-correct-service-type-for-execute-service-by-name configuration setting has been introduced for backwards compatibility. You can add the setting to the 100Custom.xml files and use it to intentionally permit authenticated users to invoke internal service types. - Resolving user mismatch between the user information cache and the database
When the user is mismatched between the user information cache and the database, the user information cache holds references to users that are not in the database. If the user information in the cache is used for further database operations, a database transaction that follows the mismatch might fail with a constraint violation exception. To prevent this mismatch, you can add the user-info-cache-block-period configuration setting to the 100Custom.xml files. Using this setting, you can specify a period of time during which the cache entry is considered invalid and is not used (if it has not been confirmed by at least one successful read from the database). - Controlling administrator access to task instance data
By default, administrators are permitted to obtain and view task instance data regardless of whether they own the associated tasks. However, you can control administrator access to task instance data by adding the authorization-enabled-for-admins-to-get-set-task-data configuration setting to the 100Custom.xml files. - Restricting Inspector actions for online Process Servers
You can limit Process Designer Inspector actions for specific Process Servers or environment types. - Specifying an absolute URL for the TWManagedFile JavaScript API
By default, a relative URL is used for the .url property of the TWManagedFile JavaScript API. However, if necessary, you can use an absolute URL instead by adding a configuration setting in the 100Custom.xml file. - Specifying a timeout for a web service
The timeout property is used to specify the amount of time to wait for a web service response when invoking an outbound request. - Controlling warning limits for variable sizes of BPDs and service flows
In IBM BPM, there are some settings in configuration files that specify warning limits for the variable sizes of business process definitions (BPDs) and service flows. If a warning limit is equaled or exceeded for any of the settings, a warning message is logged to the SystemOut.log file. You can, however, override the default warning limits for the settings by adding the settings to the 100Custom.xml file. By overriding the default warning limits of the settings, you can find the BPD, process and activity, or service flow and step name, which is causing large arrays or large values or that is using a large number of variables. This enables you to troubleshoot out-of-memory exceptions on servers. - Detecting infinite loops in services and process instances
There are a set of loop detection properties that you can use to detect and handle infinite loops in services and process instances. - Modifying the stateful session bean cache
To modify the default cache timeout, edit the session-bean-cache-expiry-timeout value in the appropriate 100Custom.xml file to your topology. - Optimizing group membership retrieval during user login
Use WebSphere native configuration to optimize group retrieval for users and their group memberships between Lightweight Directory Access Protocol (LDAP) and the IBM BPM database. - Disabling updates for user names with trailing blanks
IBM Business Process Manager (BPM) considers user names that contain trailing blanks and their corresponding entries without the trailing blanks in the Lightweight Directory Access Protocol (LDAP) registry identical. By default IBM BPM updates the information for user names with trailing blanks in its database with their LDAP counterparts during synchronization. This is also the recommended behavior. However, for compatibility reasons you can disable updates for user names with trailing blanks. - Configuring the file size of uploaded content management documents
To avoid resource depletion (such as disk space and network throughput) by uploading very large files or files that are not required for the defined use cases, the system administrator can restrict the file size of the uploaded content management documents for the document list coach view. - Hiding frequent logging of error messages
The log-tracked-loudly configuration property in the appropriate 100Custom.xml files to your topology enables or disables logging of frequent error messages. 
Hiding warnings about missing groups
The warn-of-membership-referring-to-missing-group configuration property in the appropriate 100Custom.xml files in your topology enables or disables the logging of warnings about missing groups.- Specifying that members of a group member cache are retrieved from the database
Use the group-member-cache-source setting in the appropriate 100Custom.xml file in your topology to specify that members of a group are retrieved from the IBM Business Process Manager database only, and not from a user registry, for example LDAP. This setting significantly increases the speed at which group members are loaded into the cache and reduces the number of cache reloads. - Configuring additional system lane users
You can update your configuration settings to add, replace, or remove additional system lane users in the IBM Business Process Manager environment. - Changing the block size allocation for primary keys
When instances, tasks and other objects are created in IBM BPM, a primary key is generated. Primary keys are internal to IBM BPM and they are created as needed. The default block size allocation for a primary key is 50 and this value should not need to be changed in most environments. However, in high-volume environments, it is possible that you might need to increase the block size allocation for a primary key. This is accomplished by specifying the <pri-key-block-size> configuration setting in the appropriate 100Custom.xml file to your topology. - Enforcing the scope of private variables when using the executeServiceByName JavaScript API
When you use the executeServiceByName JavaScript API, you can provide a map with input variables. If an input variable name is identical to the name of a private variable of the called service, the API overwrites the value of the private variable. To preserve the application data of the called service, you can enforce the scope of the private variables by configuring a property in the appropriate 100Custom.xml file in your topology. - Handling an IllegalDataException for invalid XML during serialization
When you convert a string that contains ASCII control characters to XML, you receive an IllegalDataException during serialization. You can add a configurable property the appropriate 100Custom.xml file in your topology to automatically make strings XML-compliant. - Extending the maximum number of characters in tracked performance data
By default, the maximum number of characters that is allowed in performance data (variable fields that are tracked to the Performance Data Warehouse for reporting purposes) is 64. You can extend this limit by making the following changes to the property file for your IBM Business Process Manager Performance Data Warehouse. The changes that you make to this property affect the column sizes in the tracking group tables. - callService configuration for running different service types
IBM Process Portal uses the callService servlet and the tw.coach.callService JavaScript API to invoke services. This servlet is configured to run Ajax services by default. If you have custom client applications that rely on callService to call service types other than Ajax services, add a configuration property to the 100Custom.xml file in IBM Process Center or IBM Process Server to specify the whitelist of callable services. - Displaying detailed error messages generated by custom applications
To display detailed error messages that are generated by custom application code instead of a generic error message, add the display-debug-error-messages configuration setting in the appropriate 100Custom.xml file in your topology. - Controlling the number of items displayed in the Event Manager monitor
You can control how many items are displayed in the Event Manager page of the Process Admin Console by configuring the event-job-threshold setting in the appropriate 100Custom.xml file in your topology. - Defining validation behavior for business objects
In IBM Process Designer, you can create business object types that define restrictions on the simple types. For example, for a business object type that has Decimal as the base type, you can specify the precision and scale settings. These settings are validated in heritage coaches and when you set a variable value in a process or service. Through various settings, you can customize the validation to behave like the totalDigits and fractionDigits constraining facets as defined by the XML Schema. - Configuring proxy settings
If you use a proxy server for internet connections, you can configure IBM Business Process Manager so that it uses the appropriate server and protocol for internet connections, such as when connecting to Blueworks Live. - Retrieving a user's full name in an easier-to-read format
When you use federated repositories, you can configure the user property in the 100Custom.xml to use the displayName attribute so that the output is easier to read. - Specifying a default time zone for work schedules
In IBM Process Portal, the time zone that you specify for work schedules for users or for when activities are due is used to calculate due dates for processes and activities. If you do not specify the time zone, Central Standard Time zone is used as the default. However, you can apply a custom default value in the 100Custom.xml files in IBM Process Server, Process Center, or both environments.