Observe that when the end user does something stupid with triggers and flows (DOES HAPPEN) the whole flow system hangs , and it is hard to remove the flow causing the error.
2 Proposals :
ONE: I observe that this is much less likely if flow is set to only run last and abort earlier. But this is not the default selection , default is run all in parallell. Changing default make system more stable ?
TWO: My runaway flows typically says something like
Executed : 49 Failed : 0 Nodes : 6 Triggers : 1 Active triggers : 1 Active flows 0
Started at : 2022-11-12 11:55:51 Duration : 9000
Possible to have a GLOBAL VARIABLE FLOW_TIMEOUT that terminates flows, ans set them to stopped when Duration is > FLOW_TIMEOUT ?