description
The dependent node is used when the dependent task is executed periodically. In the original design, if the dependent node starts execution first, but the process instance of a task does not exist, the dependent node will be directly judged as a failure. Now, with the addition of new features, dependent nodes can choose to wait for a period of time before starting execution.
To realize the idea, this feature adds a new timeout data structure, which mainly contains two attribute members: timeout and check interval. When the dependent node is running, if this function is turned on, it will check whether the pre-process instances have started at regular intervals according to the check interval. If all pre-process instances have started within the timeout period, then run directly. Otherwise, the timeout fails.
discuss mail list
Discuss on new feature for dependent node 延时等待方案讨论