Introduce workflow primitives
Introduce workflow primitives like failure handling and retry mechanism
Failure handling
Retry
Consider existing process modeling steps by various workflow management tools
We need to be able to define a sub-workflow for failure handling. In addition to that, we should be able to specify how to retry should be done. The retry can be immediate, with delay or with an exponential backoff in the initial implementation. In addition to that failure, the workflow should be able to raise an event. The key distinction between an event and a step is that event does not wait for a response coming back from an event. Although event failure should be cared about and be handled as well. If the event trigger fails, an error should be captured in the service request output.
Describe alternatives you've considered
None.
Additional context
We assume that events can be asynchronous or synchronous and can be thrown using any of the support communication protocols.
Introduce workflow primitives Introduce workflow primitives like failure handling and retry mechanism
Consider existing process modeling steps by various workflow management tools We need to be able to define a sub-workflow for failure handling. In addition to that, we should be able to specify how to retry should be done. The retry can be immediate, with delay or with an exponential backoff in the initial implementation. In addition to that failure, the workflow should be able to raise an event. The key distinction between an event and a step is that event does not wait for a response coming back from an event. Although event failure should be cared about and be handled as well. If the event trigger fails, an error should be captured in the service request output.
Describe alternatives you've considered None.
Additional context We assume that events can be asynchronous or synchronous and can be thrown using any of the support communication protocols.