Hi,
In my process model, I have a user input task (assigned to group A) then some script tasks and then I have an AND gate with two workflows coming off of it. Both workflows after the AND gate have user input tasks assigned to different groups, one being group A and one group B.
After the first user input task is completed by group A, then I would like a group A to be directly transitioned to the user task after the AND gate. I have the work flow path from the AND gate to the user input task assigned to group A activity chained but this doesn't do the direct transition.
The other user input task, assigned to group B, after the AND gate doesn't need to directly transition.
Any help is welcomed - thanks in advance!
Discussion posts and replies are publicly visible
If you are assigning to the different groups better you can have to use XOR gateway. Visual approach is better reach to give the proper answer.
Please post a screenshot of your process model configuration. If I'm assuming correctly that you're saying you want the "group A" task after the AMD gate to be chained into directly by the user who completed the prior step, this should be fine as long as your configuration is proper.
Here is the configuration. When I took away the Group B user input task, it still did not directly chain from the Group A user task to the other Group A user task.
I just tried it in my own simple example and it seems to work fine for me - the user who completes the first task chains straight into the subsequent chained task (even though it's assigned to a group).
What exactly is happening when you try?
The only thing I can think of is there is a timeout happening.
To update my original PM configuration: if there are many sub-processes (total, about 45 nodes) in between the first user task and the second group A user task, is there a timeout with activity chaining?
Yes - Appian's chaining limit is 50 by default. It counts every flow line (and some you aren't expecting), so if you think there are 45 nodes, you're probably exceeding the 50 by quite a bit.
It's really an extremely bad idea to have a situation where you need to exceed (or even, really, get close to) that number - there are alternate design approaches you should consider in this case, like off-shifting some of the subprocess functionality into asynchronous subprocess calls, etc.