Open tj opened 13 years ago
Was going to add this issue, but I see you've added it yourself. We've been running cluster in production for quite a while and those sockets can really pile up.
it's a bit of a tough call, because if you defer the exit without some kind of timeout whatever task(s) may never finish, which is definitely far from ideal
currently the events fire, but they do not allow the shutdown to halt until cleanup is complete, removing files, sockets etc