IBM BPM, V8.0.1, All platforms > Administer the IT infrastructure > Manage events > Preventing multiple failed events when a service is unavailable

Storing events

You can set up your applications so that, when a runtime error occurs during application processing, the recovery service stores the events (specifically, asynchronous requests) for later submission instead of repeatedly generating failed events.

For example, if you are deploying an application that calls a web service and that web service is unavailable, a failed event is generated and then subsequent associated asynchronous requests are stored. Then, when the service becomes available, you can forward stored asynchronous requests for processing.

• How the store-and-forward feature works
  To enable the store-and-forward feature, you set the store-and-forward qualifier on the component that will be invoked asynchronously. If that component then calls a service that is unavailable, a failed event is created and a service unavailable exception is generated. Once a failed event triggers the store function, all subsequent asynchronous requests intended for the component are stopped and stored. When the service becomes available again, you can resubmit the failed events that triggered the store and set the Store and Foward widget to forward stored events for processing.
• Store-and-forward feature set up
  Consider the following information when implementing store-and-forward processing.

Preventing multiple failed events when a service is unavailable