Application Server Failover

If the application server terminates abruptly, external systems are not affected due to the use of messaging and once the system is reachable again, any workflows that were running at the time of crashing, restart from the last known state.

In case of a graceful shutdown of the application server, the current activity of the running workflow is completed and the application then shuts down. When the application server is restarted, the workflow resumes processing normally from the point they were suspended at the shutdown time.

Note: In case of the WebSphere application server, it is a good practice to first shut down the application and then shut down the application server (to successfully save the state of currently running processes in the database).

Single Application Server Instance

If the application has a single application server instance, failover is provided - after the application server instance is restarted or is available again, workflows continue normal processing from the point they were suspended at the time of shutdown.

Multiple Application Server instances

If the application is set up with multiple application server instances acting as a cluster, in case of failure of a single application server instance, execution of a workflow proceeds due to the presence of other application server instances.

Note: All synchronous activities are repeated. Asynchronous activities are not repeatable except for the asynchronous Subflow activity.