The Ranch application with SSL listener fails after calling init:stop/0 in Erlang/OTP 17.1 and 17.3

[vicbaz]

$ git clone https://github.com/ninenines/cowboy
$ cd cowboy
$ git checkout 1.0.0
$ cd examples/ssl_hello_world
$ make
$ ./_rel/ssl_hello_world_example/bin/ssl_hello_world_example console

Erlang/OTP 17 [erts-6.2] [source-5c974be] [64-bit] [smp:4:4] [async-threads:10] [hipe] [kernel-poll:false]

Eshell V6.2  (abort with ^G)
(ssl_hello_world_example@127.0.0.1)1> application:which_applications().
[{ssl_hello_world,"Cowboy Hello World example with SSL.",
                  "1"},
 {ssl,"Erlang/OTP SSL application","5.3.6"},
 {cowboy,"Small, fast, modular HTTP server.","1.0.0"},
 {cowlib,"Support library for manipulating Web protocols.",
         "1.0.0"},
 {public_key,"Public key infrastructure","0.22.1"},
 {crypto,"CRYPTO","3.4.1"},
 {ranch,"Socket acceptor pool for TCP protocols.","1.0.0"},
 {asn1,"The Erlang ASN1 compiler version 3.0.2","3.0.2"},
 {stdlib,"ERTS  CXC 138 10","2.2"},
 {kernel,"ERTS  CXC 138 10","3.0.3"}]
(ssl_hello_world_example@127.0.0.1)2> init:stop().
ok
(ssl_hello_world_example@127.0.0.1)3>
=ERROR REPORT==== 29-Sep-2014::16:49:15 ===
Error in process <0.69.0> on node 'ssl_hello_world_example@127.0.0.1' with exit value: {{case_clause,{error,closed}},[{ranch_acceptor,loop,3,[{file,"src/ranch_acceptor.erl"},{line,28}]}]}
...
=ERROR REPORT==== 29-Sep-2014::16:49:15 ===
Error in process <0.164.0> on node 'ssl_hello_world_example@127.0.0.1' with exit value: {{case_clause,{error,closed}},[{ranch_acceptor,loop,3,[{file,"src/ranch_acceptor.erl"},{line,28}]}]}

The listener socket has been closed. The ranch_acceptors_sup shuts down the ranch application because of the maximum number of restarts. This forces all other applications to be also terminated. But the entire Erlang node remains running.

What is the purpose of this guard? As mentioned in the erlang-questions mailing list, the listener socket can be closed before the ranch application has finished its shutdown. And this behaviour is considered normal.

[chaehb]

I have struggled the same issue after erlang/otp 17.1.

[essen]

If the listening socket got closed, we need to reopen it, therefore we crash here so the error bubbles up the supervision tree until the code that reopens the socket.

The relevant mailing list discussion can be found at http://erlang.org/pipermail/erlang-questions/2014-October/081183.html

This code should not be triggered on shutdown as these processes are brutally killed.

The issue comes from the application start/stop order. If you make Ranch depend on SSL, everything works correctly. What I am unsure of is what has changed that makes code that worked before 17.1 to suddenly stop working, though.

[vicbaz]

Thank you for the clarification. I think this issue can be closed.

[essen]

I will keep it open for now as I need to investigate further, so this will be my reminder. Thanks for the report!

[MarkNijhof]

Hi @essen ,
How do I make Ranch depend on SSL to it prevents the crash on stop?

[vicbaz]

sed is your friend.

I do this:

include erlang.mk

deps::
    sed -i 's/\(stdlib\)$$/\1,\n\t\tssl/' deps/ranch/ebin/ranch.app

[MarkNijhof]

Awesome, thanks for that tip

[justnoxx]

Hi, after erlang upgrade to r18 I have this issue again. Proposed solution with sed has no effect.

[justnoxx]

Fixed sed command:

sed -i 's/\(stdlib\)/\1,\t\tssl/' deps/ranch/ebin/ranch.app

[essen]

So this is fixed in my local copy. Push later and will be in 1.3.0. Closing, thanks!