[erlang-questions] Can not kill the registered process

[Harit Himanshu]

you are right Felix, exit/1 did not help.

Instead I tried timer:kill_after() and this seems to work as expected. I
made no changes to code, only changed how to kill the registered process
and it started to work


1> c(error_handling).

{ok,error_handling}

2> error_handling:monitor_runner().

started new process <0.40.0> and reference #Ref<0.0.0.77>.

<0.39.0>

I am running.

I am running.

I am running.

I am running.

3> timer:kill_after(3000, whereis(runner)).

{ok,{1421188290386972,#Ref<0.0.0.88>}}

I am running.

runner got killed, restarting.

started new process <0.45.0> and reference #Ref<0.0.0.95>.

I am running.

I am running.

I am running.

4> timer:kill_after(3000, whereis(runner)).

{ok,{1421188310002128,#Ref<0.0.0.101>}}

runner got killed, restarting.

started new process <0.48.0> and reference #Ref<0.0.0.107>.

I am running.

I am running.

5>

Thank you
+ Harit Himanshu

On Tue, Jan 13, 2015 at 2:31 PM, Felix Gallo <felixgallo@REDACTED> wrote:

> exit/1 doesn't do what you think it does.  Maybe try exit/2?
>
> On Tue, Jan 13, 2015 at 2:23 PM, Harit Himanshu <
> harit.subscriptions@REDACTED> wrote:
>
>> The problem I am trying while learning from exercises from Joe's book is
>>
>> Write a function that creates a registered process that writes out "I’m
>> still running" every five seconds. Write a function that monitors this
>> process and restarts it if it dies. Start the global process and the
>> monitor process. Kill the global process and check that it has been
>> restarted by the monitor process.
>>
>> So I created the runner and monitor_runner as
>>
>> runner() ->
>>   receive
>>   after 5 * 1000 ->
>>     io:format("I am running.~n"),
>>     runner()
>>   end.
>>
>> monitor_runner() ->
>>   spawn(fun() ->
>>     {Pid, Ref} = spawn_monitor(error_handling, runner, []),
>>     io:format("started new process ~p and reference ~p.~n", [Pid, Ref]),
>>     register(runner, Pid),
>>     receive
>>       {'DOWN', Ref, process, Pid, _Why} ->
>>         io:format("runner got killed, restarting"),
>>         monitor_runner();
>>       X -> X
>>     end
>>   end).
>>
>>
>> When I run this, I get
>>
>> 1> c(error_handling).
>>
>> {ok,error_handling}
>>
>> 2> Pid = error_handling:monitor_runner().
>>
>> started new process <0.40.0> and reference #Ref<0.0.0.77>.
>>
>> <0.39.0>
>>
>> I am running.
>>
>> 3>
>>
>> 3>
>>
>> I am running.
>>
>> 3> Pid.
>>
>> <0.39.0>
>>
>> I am running.
>>
>> I am running.
>>
>> I am running.
>>
>> I am running.
>>
>> I am running.
>>
>> 4> whereis(runner).
>>
>> <0.40.0>
>>
>> I am running.
>>
>> 6> exit(whereis(runner)).
>>
>> ** exception exit: <0.40.0>
>>
>> I am running.
>>
>> I am running.
>>
>> I am running.
>>
>> I am running.
>>
>> I am running.
>>
>> I am running.
>>
>> I am running.
>>
>> 7>
>>
>> *Problem?*
>>  The runner is still running.
>> 1.) Why is the process not getting killed?
>> 2.) Why is it not caught by {'DOWN'....}
>>
>> What am I missing here?
>>
>> Thanks
>> + Harit
>>
>> _______________________________________________
>> erlang-questions mailing list
>> erlang-questions@REDACTED
>> http://erlang.org/mailman/listinfo/erlang-questions
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20150113/d100deac/attachment.htm>