Assume there is a function like: def sample do lab_pid = self() spawn_link(fn -> spawn_link(fn -> Process.flag(:trap_exit, true) receive do {:EXIT, _pid, :normal} -> send(lab_pid, :done) end end) :timer.sleep(100) end) receive do msg ->…

I wouldn’t call it a deadlock, but rather a race condition. Too see why it happens, let’s first name these processes: def sample do # process A lab_pid = self() spawn_link(fn -> # process B spawn_link(fn -> # process C Process.flag(:trap_exit, true) receive do …

You are right! A simpler way (for me) is informing the parent that the child is started/ready: def sample do lab_pid = self() spawn_link(fn -> parent_pid = self() spawn_link(fn -> Process.flag(:trap_exit, true) send(parent_pid, :started) receive do {:EXIT, _…

[image] dc0d: A simpler way (for me) is informing the parent that the child is started/ready This is what :proc_lib.start_link and :proc_lib.init_ack do under the hood too :slight_smile: See here and here .

This is also a good example as to why monitors are often better then links+traps. :slight_smile:

Since here the child is trapping the exit of parent (to do some cleaups for example) wouldn’t the situation be the same? For example the parent process can exit before the Process.monitor(parent_pid) expression executes inside the child process.

[image] dc0d: For example the parent process can exit before the Process.monitor(parent_pid) expression executes inside the child process. That’s why a parent supervisor of both should ensure that the monitoring process always outlives the children.

[image] dc0d: For example the parent process can exit before the Process.monitor(parent_pid) expression executes inside the child process. In this case you would still receive a :DOWN message with an exit reason set to :noproc, so monitor would in fact work without additional synchronism. Tha…

[image] sasajuric: In this case you would still receive a :DOWN message with an exit reason set to :noproc , so monitor would in fact work without additional synchronism. That said, if we’re talking about parent-child relationship, the usual approach in OTP is to use the exit signals to do the c…

Stale/Deadlock Process

Questions & Help Questions

OvermindDL1 November 14, 2018, 6:35pm 7

That’s why a parent supervisor of both should ensure that the monitoring process always outlives the children.