Monitor Rules

You can define algorithms to monitor a business entity while it is in a given state. This type of logic is typically used to check if the conditions necessary to transition the entity to another state exist (and, if so, transition it). For example, transition an entity to the Canceled state if it's been in the Error state too long. Another common use is to perform ancillary work while an entity is in a given state. For example, update statistics held on the object while it's in the Active state .

Monitor algorithms are invoked when a business entity first enters a state and periodically after that in batch. You have the option to defer the monitoring of a specific state until a specific monitoring batch job runs. You do so by associating the state with a specific monitoring process. In this case the system will only execute the monitoring rules of this state when that specific batch process runs. This may be useful for example in a market-messaging world where you do not want an inbound message processed when it is received; rather, you want to wait until a later point in time (maybe at the end of the day).

A monitor algorithm can carry out any business logic. In addition it can optionally tell the system to do either of the following:

• Stop monitoring and transition to another state. The system will not call any further monitoring algorithm plugged in on the state and attempt to transition the entity to the requested new state.
• Stop monitoring. Same as above except that no transition takes place. You may want to use this option to prevent transitions while some condition is true.

If none of the above is requested the system keeps executing subsequent monitoring algorithms.