Does anyone know - is it an intentional design decision?
I’m trying to link it to a process that’s not its direct parent/child, so it would exit when the other process exits, but of course at the moment it doesn’t work. I don’t care so much about my use-case per-se (I can figure out a solution), but I’m in general wondering if there’s a good reason for how DynamicSupervisor is implemented - i.e. maybe I’m doing something that goes against first principles?
Ultimately I’m interested in shutting down the supervisor when that process exits. And vice versa - I want the process to shut down when the supervisor exits.
Establishing a link between them would’ve been the easiest option.
Make that proces being supervised by that supervisor instead, and then make it :significant (though I do not know if DynamicSupervisor supports that, but if that process is known to be there, then why it is part of DynamicSupervisor?).