Hi Martin,
On Tue, Jan 4, 2011 at 1:53 AM, Martin Sustrik <sust...@250bpm.com> wrote:
> Hi Dhammika,
>
>> It's actually in zmq.
>> Terminated pipe writer gets a "activate" command. But load balancer
>> doesn't decrement active pipe count in terminate() call.
>>
>> 55 void zmq::lb_t::terminate ()
>> 56 {
>> 57 zmq_assert (!terminating);
>> 58 terminating = true;
>> 59
>> 60 sink->register_term_acks (pipes.size ());
>> 61 for (pipes_t::size_type i = 0; i != pipes.size (); i++)
>> 62 pipes [i]->terminate ();
>> 63 }
>
> Hm. Termination is an async process. So, at this point termination of all
> the pipes is started, but it will take some time to finish. As the time
> progresses, individual pipes will unregister themselves from the
> load-balancer.
>
Yes, terminate() just initiates shutdown process.
But on "activate_writer" command pipe writer tries to move the pipe
back into active list, even when it's in terminating state.
301 void zmq::writer_t::process_activate_writer (uint64_t msgs_read_)
302 {
340 // If the writer was non-active before, let's make it active
341 // (available for writing messages to).
342 if (!active) {
343 active = true;
344 zmq_assert (sink);
345 sink->activated (this);
346 }
347 }
But load balancer hasn't removed the pipe from its active list, so
"active" index in following swap call is off.
80 void zmq::lb_t::activated (writer_t *pipe_)
81 {
82 // Move the pipe to the list of active pipes.
83 pipes.swap (pipes.index (pipe_), active);
84 active++;
85 }
Dhammika
_______________________________________________
zeromq-dev mailing list
zeromq-dev@lists.zeromq.org
http://lists.zeromq.org/mailman/listinfo/zeromq-dev
