[rabbitmq-discuss] Creating an auth plugin (Kerberos)

Wed Dec 5 13:42:04 GMT 2012

On Wed, 2012-12-05 at 14:12:21 +0100, Simon Lundström wrote:
> > One thing I would suggest is that if you do hit `after 5000` and time out whilst waiting for the external program to complete, it *might* be a good idea to close the port so as not to leak resources. Check the open_port documentation for details.
> 
> Ah, I didn't read the eof documentation good enough it seems.
> 
> "The port will not be closed at the end of the file and produce an exit signal."
> 
> Thank you, good catch!

Hm, no matter where I put the port_close/1 call (I've tried kinit/2,
kinit/3 and loop/2) and even if I receive the {'EXIT'} message in my
loop/2 I still get:

=CRASH REPORT==== 5-Dec-2012::14:30:43 ===
  crasher:
    initial call: rabbit_reader:init/4
    pid: <0.280.0>
    registered_name: []
    exception exit: {unexpected_message,{'EXIT',#Port<0.6598>,normal}}
      in function  rabbit_reader:handle_other/3
      in call from rabbit_reader:start_connection/7
    ancestors: [<0.277.0>,rabbit_tcp_client_sup,rabbit_sup,<0.149.0>]
    messages: []
    links: [<0.277.0>]
    dictionary: []
    trap_exit: true
    status: running
    heap_size: 2584
    stack_size: 24
    reductions: 6594
  neighbours:

=SUPERVISOR REPORT==== 5-Dec-2012::14:30:43 ===
     Supervisor: {<0.277.0>,rabbit_connection_sup}
     Context:    child_terminated
     Reason:     {unexpected_message,{'EXIT',#Port<0.6598>,normal}}
     Offender:   [{pid,<0.280.0>},
                  {name,reader},
                  {mfa,{rabbit_reader,start_link,
                                      [<0.279.0>,<0.278.0>,
                                       #Fun<rabbit_heartbeat.2.20850803>]}},
                  {restart_type,intrinsic},
                  {shutdown,4294967295},
                  {child_type,worker}]

=SUPERVISOR REPORT==== 5-Dec-2012::14:30:43 ===
     Supervisor: {<0.277.0>,rabbit_connection_sup}
     Context:    shutdown
     Reason:     reached_max_restart_intensity
     Offender:   [{pid,<0.280.0>},
                  {name,reader},
                  {mfa,{rabbit_reader,start_link,
                                      [<0.279.0>,<0.278.0>,
                                       #Fun<rabbit_heartbeat.2.20850803>]}},
                  {restart_type,intrinsic},
                  {shutdown,4294967295},
                  {child_type,worker}]

Must I use my own supervisor (which I have removed before)?
How can I check if the Port leakage actually exists? How can I see what
ports are open in my process?

Thanks,
- Simon