pids and ets ids leaks

[Veselov Andrey]

3> f(Pid1), f(Pid2), [Pid1, Pid2] = [erlang:spawn_link(fun() -> receive _ -> ok = folsom_metrics:new_histogram(<<"sorry">>, slide) end end) || _ <- lists:seq(1, 2)].
[<0.45.0>,<0.46.0>]
4>  Pid1 ! Pid2 ! sorry.
sorry
5> supervisor:which_children(folsom_sample_slide_sup).
[{undefined,<0.49.0>,worker,[folsom_sample_slide_server]},
 {undefined,<0.50.0>,worker,[folsom_sample_slide_server]}]
7> ets:tab2list(folsom_histograms).
[{<<"sorry">>,
  {histogram,slide,{slide,1028,20506,<0.50.0>}}}]

I expected that there would be only one child of folsom_sample_slide_sup. So the second child and its ets are leaked.

supervisor:count_children(folsom_sample_slide_sup) =/= erlang:length(ets:tab2list(folsom_histograms)).

metrics with the same leaks:

• history
• spiral
• duration

[Joe Williams]

Folsom has moved, please resubmit your issue at https://github.com/folsom-project Thanks!