Changing message flow behavior
You can change the behavior that is taken by your message flows to process messages in different ways and at different times.
About this task
If one or more of the default behaviors that are described in Default message flow behavior is not sufficient or appropriate for your message processing, you can change some characteristics of that behavior at different times during the design, development, and production cycles:
Behavior you can change when you design your message flows
About this task
- Transactional support
- You can configure your message flows to handle your messages, and other data, in transactions. These options are described in Configuring transactionality for message flows.
- Error handling
- The integration node provides an initial level of error processing
for all message flows. If you require further support in your message
flows, you can add your own processing. You can learn more about these
options in Handling errors in message flows.
The nodes that support some protocols provide extra default error handling; this support is described, where relevant, in the sections in Connecting client applications.
- Data conversion
- If you are exchanging messages between unlike systems, you can update integration node or WebSphere® MQ configuration, or design and supply your own conversion procedures. Options are described in Configuring message flows for data conversion.
- User-defined properties
- You can create user-defined properties for your message flows to associate values with those message flows. You can then configure the nodes in your message flows to access those properties and their values by coding ESQL or Java™ programs. Read more about user-defined properties, and see how to create them in the Message Flow editor.
- Promoted properties
- You can promote some message node properties to the level of the message flow in which the node is included. The advantages of this technique are described in Promoted properties; see Defining a promoted property for details of how to use these properties.
Behavior you can change when you deploy your message flows
About this task
- Configurable properties and BAR overrides
- Some of the properties on message flow nodes are configurable;
that is, you can change their values when you deploy the message flow.
By using this option, you can change some characteristics of a deployed
message flow without changing the message flow definitions. For example,
you can update queue manager and data source information in the BAR
file when you deploy it.
You can change these values by using the IBM Integration Toolkit, or the mqsiapplybaroverride command.
- The execution and threading model
- You can increase the number of threads that your message flows use to reduce the time in which your messages are handled. You can also deploy multiple copies of a message flow to one or more integration nodes. For more information, see Optimizing message flow throughput.
Behavior you can change at any time
About this task
- Operational policies
- An operational policy enables you to define a common approach to controlling certain aspects of message flow behavior, including flow rate, and particular node properties such as connection credentials. You can update an operational policy at any time in the solution lifecycle. For more information about operational policies, see Operational policy.
Behavior you can change during message flow processing
About this task
- Configurable services
- By using a configurable service, you can create and configure
properties that relate to external services called by the integration
node from your message flows. A number of configurable services are
supplied with the integration node; you can modify or add to these
default services, and you can create your own services.
Learn more about this option in Configurable services.
- Local environment overrides
- You can configure some nodes to include your own processing; by coding ESQL, Java, or maps, you can modify the contents of the local environment tree within a message. Some fields in the local environment are used by nodes to determine how the message is processed, therefore by changing the tree contents, you can influence the behavior of subsequent nodes in the message flow. For more information about this option, see Transforming and enriching messages.