From: Eric Wong Date: 2017-08-09T23:33:17+00:00 Subject: [ruby-core:82319] Re: [Ruby trunk Bug#13794] Infinite loop of sched_yield charlie@atech.media wrote: > Bug #13794: Infinite loop of sched_yield > https://bugs.ruby-lang.org/issues/13794 > ---------------------------------------- > I have been encountering an issue with processes hanging in an infinite loop of calling sched_yield(). The looping code can be found at https://github.com/ruby/ruby/blob/v2_3_4/thread_pthread.c#L1663 > > while (ATOMIC_CAS(timer_thread_pipe.writing, (rb_atomic_t)0, 0)) { > native_thread_yield(); > } > > It is my belief that by some mechanism I have not been able to identify, timer_thread_pipe.writing is incremented but it never decremented, causing this loop to run infinitely. > > I am not able to create a reproducible test case, however this issue occurs regularly in my production application. I have attached backtraces and thread lists from 2 processes exhibiting this behaviour. gdb confirms that timer_thread_pipe.writing = 1 in these processes. Can you also check the value of timer_thread_pipe.owner_process? > I believe one possibility of the cause is that rb_thread_wakeup_timer_thread() or rb_thread_wakeup_timer_thread_low() is called, and before it returns, another thread calls fork(), leaving the value of timer_thread_pipe.writing incremented, but leaving behind the thread that would normally decrement it. That is a likely possibility. > If this is correct, one solution would be to reset timer_thread_pipe.writing to 0 in native_reset_timer_thread() immediately after a fork. How about checking owner_process before incrementing? Can you try the following patch to check owner_process? https://80x24.org/spew/20170809232533.14932-1-e@80x24.org/raw timer_thread_pipe.writing was introduced in August 2015 with r51576, so this bug would definitely be my fault. > Other examples of similar bugs being reported: > https://github.com/resque/resque/issues/578 > https://github.com/zk-ruby/zk/issues/50 That also means these bugs from 2012 are from other causes. Thanks again for this report. Unsubscribe: