Parallel Gateway Limitation Example

Please let me know how Parallel gateway can receive tokens without coming on every incoming flow

Limitation

Note that in Camunda’s implementation of the parallel gateway, the gateway triggers as soon as the following holds: The number of arrived tokens is equal to the number of incoming sequence flows. It is not required that a token arrives on every incoming flow.

No - it doesn’t matter where the tokens come from, only the number of tokens matters.

Thanks for update. But, How tokens can come without following incoming flow. Please share the example.

Hi @Ganesh_Gupta,

Below model is an example

Where we have 3 parallel user tasks (A, B and C). Two of them followed by exclusive gateway.
Assuming below scenario

  • Task A is completed while the 2 others are still pending. (a token will get out from the exclusive gateway and arrive at the 2nd parallel gateway)

  • Now Task B is completed while C is still pending. (a new token will get out from the exclusive gateway and now 2 tokens have arrived at the 2nd parallel gateway)

  • Now the 2nd parallel gateway will trigger as the number of arrived tokens (2) is equal to the number of incoming sequence flows (2) despite the fact that Task C is still pending.