Hi all,<div><br></div><div>Trying to send again under a new account...</div><div><br></div><div>Cheers,</div><div><br></div><div>Pete<br><br><div class="gmail_quote">---------- Forwarded message ----------<br>From: <b class="gmail_sendername">Peter Membrey</b> <span dir="ltr"><<a href="mailto:peter@membrey.hk">peter@membrey.hk</a>></span><br>
Date: 24 November 2012 21:57<br>Subject: Re: [erlang-bugs] VM locks up on write to socket (and now it seems to file too)<br>To: Patrik Nyblom <<a href="mailto:pan@erlang.org">pan@erlang.org</a>><br>Cc: <a href="mailto:erlang-bugs@erlang.org">erlang-bugs@erlang.org</a><br>
<br><br>Hi guys,<br>
<br>
Thanks for getting back in touch so quickly!<br>
<br>
I did do an lsof on the process and I can confirm that it was<br>
definitely a socket. However by that time the application it had been<br>
trying to send to had been killed. When I checked the sockets were<br>
showing as waiting to close. Unfortunately I didn't think to do an<br>
lsof until after the apps had been shut down. I was hoping the VM<br>
would recover if I killed the app that had upset it. However even<br>
after all the apps connected had been shut down, the issue didn't<br>
resolve.<br>
<br>
The application receives requests from a client, which contains two<br>
data items. The stream ID and a timestamp. Both are encoded as big<br>
integer unsigned numbers. The server then looks through the file<br>
referenced by the stream ID and uses the timestamp as an index. The<br>
file format is currently really simple, in the form of:<br>
<br>
<Timestamp:64/native-integer,Length:32/native-integer,Data:Length/binary>><br>
<br>
There is an index file that provides an offset into the file based on<br>
time stamp, but basically it opens the file, and reads sequentially<br>
through it until it finds the timestamps that it cares about. In this<br>
case it reads all data with a greater timestamp until the end of the<br>
file is reached. It's possible the client is sending an incorrect<br>
timestamp, and maybe too much data is being read. However the loop is<br>
very primitive - it reads all the data in one go before passing it<br>
back to the protocol handler to send down the socket; so by that time<br>
even though the response is technically incorrect and the app has<br>
failed, it should still not cause the VM any issues.<br>
<br>
The data is polled every 10 seconds by the client app so I would not<br>
expect there to be 2GB of new data to send. I'm afraid my C skills are<br>
somewhat limited, so I'm not sure how to put together a sample app to<br>
try out writev. The platform is 64bit CentOS 6.3 (equivalent to RHEL<br>
6.3) so I'm not expecting any strange or weird behaviour from the OS<br>
level but of course I could be completely wrong there. The OS is<br>
running directly on hardware, so there's no VM layer to worry about.<br>
<br>
Hope this might offer some additional clues…<br>
<br>
Thanks again!<br>
<br>
Kind Regards,<br>
<br>
Peter Membrey<br>
<div class="HOEnZb"><div class="h5"><br>
<br>
<br>
On 24 November 2012 00:13, Patrik Nyblom <<a href="mailto:pan@erlang.org">pan@erlang.org</a>> wrote:<br>
> Hi again!<br>
><br>
> Could you go back to the version without the printouts and get back to the<br>
> situation where writev loops returning 0 (as in the strace)? If so, it would<br>
> be really interesting to see an 'lsof' of the beam process, to see if this<br>
> file descriptor really is open and is a socket...<br>
><br>
> The thing is that writev with a vector that is not empty, would never return<br>
> 0 for a non blocking socket. Not on any modern (i.e. not ancient) POSIX<br>
> compliant system anyway. Of course it is a *really* large item you are<br>
> trying to write there, but it should be no problem for a 64bit linux.<br>
><br>
> Also I think there is no use finding the Erlang code, I'll take that back,<br>
> It would be more interesting to see what really happens at the OS/VM level<br>
> in this case.<br>
><br>
> Cheers,<br>
> Patrik<br>
><br>
><br>
> On 11/23/2012 01:49 AM, Loďc Hoguin wrote:<br>
>><br>
>> Sending this on behalf of someone who didn't manage to get the email sent<br>
>> to this list after 2 attempts. If someone can check if he's hold up or<br>
>> something that'd be great.<br>
>><br>
>> Anyway he has a big issue so I hope I can relay the conversation reliably.<br>
>><br>
>> Thanks!<br>
>><br>
>> On 11/23/2012 01:45 AM, Peter Membrey wrote:<br>
>>><br>
>>> From: Peter Membrey <<a href="mailto:peter@membrey.hk">peter@membrey.hk</a>><br>
>>> Date: 22 November 2012 19:02<br>
>>> Subject: VM locks up on write to socket (and now it seems to file too)<br>
>>> To: <a href="mailto:erlang-bugs@erlang.org">erlang-bugs@erlang.org</a><br>
>>><br>
>>><br>
>>> Hi guys,<br>
>>><br>
>>> I wrote a simple database application called CakeDB<br>
>>> (<a href="https://github.com/pmembrey/cakedb" target="_blank">https://github.com/pmembrey/cakedb</a>) that basically spends its time<br>
>>> reading and writing files and sockets. There's very little in the way<br>
>>> of complex logic. It is running on CentOS 6.3 with all the updates<br>
>>> applied. I hit this problem on R15B02 so I rolled back to R15B01 but<br>
>>> the issue remained. Erlang was built from source.<br>
>>><br>
>>> The machine has two Intel X5690 CPUs giving 12 cores plus HT. I've<br>
>>> tried various arguments for the VM but so far nothing has prevented<br>
>>> the problem. At the moment I'm using:<br>
>>><br>
>>> +K<br>
>>> +A 6<br>
>>> +sbt tnnps<br>
>>><br>
>>> The issue I'm seeing is that one of the scheduler threads will hit<br>
>>> 100% cpu usage and the entire VM will become unresponsive. When this<br>
>>> happens, I am not able to connect via the console with attach and<br>
>>> entop is also unable to connect. I can still establish TCP connections<br>
>>> to the application, but I never receive a response. A standard kill<br>
>>> signal will cause the VM to shut down (it doesn't need -9).<br>
>>><br>
>>> Due to the pedigree of the VM I am quite willing to accept that I've<br>
>>> made a fundamental mistake in my code. I am pretty sure that the way I<br>
>>> am doing the file IO could result in some race conditions. However, my<br>
>>> poor code aside, from what I understand, I still shouldn't be able to<br>
>>> crash / deadlock the VM like this.<br>
>>><br>
>>> The issue doesn't seem to be caused by load. The app can fail when<br>
>>> it's very busy, but also when it is practically idle. I haven't been<br>
>>> able to find a trigger or any other explanation for the failure.<br>
>>><br>
>>> The thread maxing out the CPU is attempting to write data to the socket:<br>
>>><br>
>>> (gdb) bt<br>
>>> #0 0x00007f9882ab6377 in writev () from /lib64/libc.so.6<br>
>>> #1 0x000000000058a81f in tcp_inet_output (data=0x2407570,<br>
>>> event=<value optimized out>) at drivers/common/inet_drv.c:9681<br>
>>> #2 tcp_inet_drv_output (data=0x2407570, event=<value optimized out>)<br>
>>> at drivers/common/inet_drv.c:9601<br>
>>> #3 0x00000000004b773f in erts_port_task_execute (runq=0x7f98826019c0,<br>
>>> curr_port_pp=0x7f9881639338) at beam/erl_port_task.c:858<br>
>>> #4 0x00000000004afd83 in schedule (p=<value optimized out>,<br>
>>> calls=<value optimized out>) at beam/erl_process.c:6533<br>
>>> #5 0x0000000000539ca2 in process_main () at beam/beam_emu.c:1268<br>
>>> #6 0x00000000004b1279 in sched_thread_func (vesdp=0x7f9881639280) at<br>
>>> beam/erl_process.c:4834<br>
>>> #7 0x00000000005ba726 in thr_wrapper (vtwd=0x7fff6cfe2300) at<br>
>>> pthread/ethread.c:106<br>
>>> #8 0x00007f9882f78851 in start_thread () from /lib64/libpthread.so.0<br>
>>> #9 0x00007f9882abe11d in clone () from /lib64/libc.so.6<br>
>>> (gdb)<br>
>>><br>
>>> I then tried running strace on that thread and got (indefinitely):<br>
>>><br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> writev(15, [{"", 2158022464}], 1) = 0<br>
>>> ...<br>
>>><br>
>>> From what I can tell, it's trying to write data to a socket, which is<br>
>>> succeeding, but writing 0 bytes. From the earlier definitions in the<br>
>>> source file, an error condition would be signified by a negative<br>
>>> number. Any other result is the number of bytes written, in this case<br>
>>> 0. I'm not sure if this is desired behaviour or not. I've tried<br>
>>> killing the application on the other end of the socket, but it has no<br>
>>> effect on the VM.<br>
>>><br>
>>> I have enabled debugging for the inet code, so hopefully this will<br>
>>> give a little more insight. I am currently trying to reproduce the<br>
>>> condition, but as I really have no idea what causes it, it's pretty<br>
>>> much a case of wait and see.<br>
>>><br>
>>><br>
>>> **** UPDATE ****<br>
>>><br>
>>> I managed to lock up the VM again, but this time it was caused by file<br>
>>> IO,<br>
>>> probably from the debugging statements. Although it worked fine for some<br>
>>> time<br>
>>> the last entry in the file was cut off.<br>
>>><br>
>>> From GDB:<br>
>>><br>
>>> (gdb) info threads<br>
>>> 53 Thread 0x7f83e988b700 (LWP 8621) 0x00007f83ea6da54d in read ()<br>
>>> from /lib64/libpthread.so.0<br>
>>> 52 Thread 0x7f83e8c8f700 (LWP 8622) 0x00007f83ea6d743c in<br>
>>> pthread_cond_wait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0<br>
>>> 51 Thread 0x7f83e818d700 (LWP 8623) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 50 Thread 0x7f83e816b700 (LWP 8624) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 49 Thread 0x7f83e8149700 (LWP 8625) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 48 Thread 0x7f83e8127700 (LWP 8626) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 47 Thread 0x7f83e8105700 (LWP 8627) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 46 Thread 0x7f83e80e3700 (LWP 8628) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 45 Thread 0x7f83e80c1700 (LWP 8629) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 44 Thread 0x7f83e809f700 (LWP 8630) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 43 Thread 0x7f83e807d700 (LWP 8631) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 42 Thread 0x7f83e805b700 (LWP 8632) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 41 Thread 0x7f83e8039700 (LWP 8633) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 40 Thread 0x7f83e8017700 (LWP 8634) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 39 Thread 0x7f83e7ff5700 (LWP 8635) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 38 Thread 0x7f83e7fd3700 (LWP 8636) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 37 Thread 0x7f83e7fb1700 (LWP 8637) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 36 Thread 0x7f83e7f8f700 (LWP 8638) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 35 Thread 0x7f83e7f6d700 (LWP 8639) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 34 Thread 0x7f83e7f4b700 (LWP 8640) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 33 Thread 0x7f83e7f29700 (LWP 8641) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 32 Thread 0x7f83e7f07700 (LWP 8642) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 31 Thread 0x7f83e7ee5700 (LWP 8643) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 30 Thread 0x7f83e7ec3700 (LWP 8644) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 29 Thread 0x7f83e7ea1700 (LWP 8645) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 28 Thread 0x7f83e7e7f700 (LWP 8646) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 27 Thread 0x7f83d7c5a700 (LWP 8647) 0x00007f83ea6db09d in waitpid<br>
>>> () from /lib64/libpthread.so.0<br>
>>> 26 Thread 0x7f83d7c53700 (LWP 8648) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 25 Thread 0x7f83d7252700 (LWP 8649) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 24 Thread 0x7f83d6851700 (LWP 8650) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 23 Thread 0x7f83d5e50700 (LWP 8651) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 22 Thread 0x7f83d544f700 (LWP 8652) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 21 Thread 0x7f83d4a4e700 (LWP 8653) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 20 Thread 0x7f83d404d700 (LWP 8654) 0x00007f83ea20be7d in write ()<br>
>>> from /lib64/libc.so.6<br>
>>> 19 Thread 0x7f83d364c700 (LWP 8655) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 18 Thread 0x7f83d2c4b700 (LWP 8656) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 17 Thread 0x7f83d224a700 (LWP 8657) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 16 Thread 0x7f83d1849700 (LWP 8658) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 15 Thread 0x7f83d0e48700 (LWP 8659) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 14 Thread 0x7f83d0447700 (LWP 8660) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 13 Thread 0x7f83cfa46700 (LWP 8661) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 12 Thread 0x7f83cf045700 (LWP 8662) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 11 Thread 0x7f83ce644700 (LWP 8663) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 10 Thread 0x7f83cdc43700 (LWP 8664) 0x00007f83ea215ae9 in syscall<br>
>>> () from /lib64/libc.so.6<br>
>>> 9 Thread 0x7f83cd242700 (LWP 8665) 0x00007f83ea215ae9 in syscall ()<br>
>>> from /lib64/libc.so.6<br>
>>> 8 Thread 0x7f83cc841700 (LWP 8666) 0x00007f83ea215ae9 in syscall ()<br>
>>> from /lib64/libc.so.6<br>
>>> 7 Thread 0x7f83cbe40700 (LWP 8667) 0x00007f83ea215ae9 in syscall ()<br>
>>> from /lib64/libc.so.6<br>
>>> 6 Thread 0x7f83cb43f700 (LWP 8668) 0x00007f83ea215ae9 in syscall ()<br>
>>> from /lib64/libc.so.6<br>
>>> 5 Thread 0x7f83caa3e700 (LWP 8669) 0x00007f83ea215ae9 in syscall ()<br>
>>> from /lib64/libc.so.6<br>
>>> 4 Thread 0x7f83ca03d700 (LWP 8670) 0x00007f83ea215ae9 in syscall ()<br>
>>> from /lib64/libc.so.6<br>
>>> 3 Thread 0x7f83c963c700 (LWP 8671) 0x00007f83ea215ae9 in syscall ()<br>
>>> from /lib64/libc.so.6<br>
>>> 2 Thread 0x7f83c8c3b700 (LWP 8672) 0x00007f83ea215ae9 in syscall ()<br>
>>> from /lib64/libc.so.6<br>
>>> * 1 Thread 0x7f83eb3a8700 (LWP 8597) 0x00007f83ea211d03 in select ()<br>
>>> from /lib64/libc.so.6<br>
>>> (gdb)<br>
>>><br>
>>><br>
>>> (gdb) bt<br>
>>> #0 0x00007f83ea20be7d in write () from /lib64/libc.so.6<br>
>>> #1 0x00007f83ea1a2583 in _IO_new_file_write () from /lib64/libc.so.6<br>
>>> #2 0x00007f83ea1a3b35 in _IO_new_do_write () from /lib64/libc.so.6<br>
>>> #3 0x00007f83ea1a21fd in _IO_new_file_xsputn () from /lib64/libc.so.6<br>
>>> #4 0x00007f83ea17589d in vfprintf () from /lib64/libc.so.6<br>
>>> #5 0x00007f83ea18003a in printf () from /lib64/libc.so.6<br>
>>> #6 0x000000000058f0e8 in tcp_recv (desc=0x2c3d350, request_len=0) at<br>
>>> drivers/common/inet_drv.c:8976<br>
>>> #7 0x000000000058f63a in tcp_inet_input (data=0x2c3d350, event=<value<br>
>>> optimized out>) at drivers/common/inet_drv.c:9326<br>
>>> #8 tcp_inet_drv_input (data=0x2c3d350, event=<value optimized out>)<br>
>>> at drivers/common/inet_drv.c:9604<br>
>>> #9 0x00000000004b770f in erts_port_task_execute (runq=0x7f83e9d5d3c0,<br>
>>> curr_port_pp=0x7f83e8dc6e78) at beam/erl_port_task.c:851<br>
>>> #10 0x00000000004afd83 in schedule (p=<value optimized out>,<br>
>>> calls=<value optimized out>) at beam/erl_process.c:6533<br>
>>> #11 0x0000000000539ca2 in process_main () at beam/beam_emu.c:1268<br>
>>> #12 0x00000000004b1279 in sched_thread_func (vesdp=0x7f83e8dc6dc0) at<br>
>>> beam/erl_process.c:4834<br>
>>> #13 0x00000000005bb3e6 in thr_wrapper (vtwd=0x7fffe8266da0) at<br>
>>> pthread/ethread.c:106<br>
>>> #14 0x00007f83ea6d3851 in start_thread () from /lib64/libpthread.so.0<br>
>>> #15 0x00007f83ea21911d in clone () from /lib64/libc.so.6<br>
>>> (gdb)<br>
>>><br>
>>> (gdb) bt<br>
>>> #0 0x00007f83ea6da54d in read () from /lib64/libpthread.so.0<br>
>>> #1 0x0000000000554b6e in signal_dispatcher_thread_func (unused=<value<br>
>>> optimized out>) at sys/unix/sys.c:2776<br>
>>> #2 0x00000000005bb3e6 in thr_wrapper (vtwd=0x7fffe8266c80) at<br>
>>> pthread/ethread.c:106<br>
>>> #3 0x00007f83ea6d3851 in start_thread () from /lib64/libpthread.so.0<br>
>>> #4 0x00007f83ea21911d in clone () from /lib64/libc.so.6<br>
>>> (gdb)<br>
>>><br>
>>> (gdb) bt<br>
>>> #0 0x00007f83ea215ae9 in syscall () from /lib64/libc.so.6<br>
>>> #1 0x00000000005bba35 in wait__ (e=0x2989390) at<br>
>>> pthread/ethr_event.c:92<br>
>>> #2 ethr_event_wait (e=0x2989390) at pthread/ethr_event.c:218<br>
>>> #3 0x00000000004ae5bd in erts_tse_wait (fcalls=<value optimized out>,<br>
>>> esdp=0x7f83e8e2c440, rq=0x7f83e9d5e7c0) at beam/erl_threads.h:2319<br>
>>> #4 scheduler_wait (fcalls=<value optimized out>, esdp=0x7f83e8e2c440,<br>
>>> rq=0x7f83e9d5e7c0) at beam/erl_process.c:2087<br>
>>> #5 0x00000000004afb94 in schedule (p=<value optimized out>,<br>
>>> calls=<value optimized out>) at beam/erl_process.c:6467<br>
>>> #6 0x0000000000539ca2 in process_main () at beam/beam_emu.c:1268<br>
>>> #7 0x00000000004b1279 in sched_thread_func (vesdp=0x7f83e8e2c440) at<br>
>>> beam/erl_process.c:4834<br>
>>> #8 0x00000000005bb3e6 in thr_wrapper (vtwd=0x7fffe8266da0) at<br>
>>> pthread/ethread.c:106<br>
>>> #9 0x00007f83ea6d3851 in start_thread () from /lib64/libpthread.so.0<br>
>>> #10 0x00007f83ea21911d in clone () from /lib64/libc.so.6<br>
>>> (gdb)<br>
>>><br>
>>><br>
>>> (gdb) bt<br>
>>> #0 0x00007f83ea6db09d in waitpid () from /lib64/libpthread.so.0<br>
>>> #1 0x0000000000555a9f in child_waiter (unused=<value optimized out>)<br>
>>> at sys/unix/sys.c:2700<br>
>>> #2 0x00000000005bb3e6 in thr_wrapper (vtwd=0x7fffe8266d50) at<br>
>>> pthread/ethread.c:106<br>
>>> #3 0x00007f83ea6d3851 in start_thread () from /lib64/libpthread.so.0<br>
>>> #4 0x00007f83ea21911d in clone () from /lib64/libc.so.6<br>
>>> (gdb)<br>
>>><br>
>>><br>
>>> **** END UPDATE ****<br>
>>><br>
>>><br>
>>> I'm happy to provide any information I can, so please don't hesitate to<br>
>>> ask.<br>
>>><br>
>>> Thanks in advance!<br>
>>><br>
>>> Kind Regards,<br>
>>><br>
>>> Peter Membrey<br>
>>><br>
>><br>
>><br>
><br>
> _______________________________________________<br>
> erlang-bugs mailing list<br>
> <a href="mailto:erlang-bugs@erlang.org">erlang-bugs@erlang.org</a><br>
> <a href="http://erlang.org/mailman/listinfo/erlang-bugs" target="_blank">http://erlang.org/mailman/listinfo/erlang-bugs</a><br>
</div></div></div><br></div>