[zeromq-dev] zmq::zmq_abort -- what's that about?

[Benson Margulies]

Yes, I have threads. Thanks to your hint, I found a locking error and
got rid of the abort. I wish it were as simple as 'don't do that,' but
I have an existing architecture to fit into and I have not yet thought
of a way to eliminate some sharing under a lock.

On Fri, Apr 10, 2015 at 3:23 AM, Dario Meloni <mellon85 at gmail.com> wrote:
>
> On Thu, 09 Apr 2015 17:48:40 -0400, Benson Margulies wrote:
>
>> I have a core dump with the following backtrace. Any suggestions as to
>> how to make heads or tails?
>>
>> #6  0x0000000000c12869 in zmq::zmq_abort
>> (errmsg_=errmsg_ at entry=0x2136f40 "ok") at err.cpp:74 #7
>> 0x0000000000c13105 in zmq::mailbox_t::recv
>> (this=this at entry=0x7f18be60e400, cmd_=cmd_ at entry=0x7f18b9ffdea0,
>>     timeout_=<optimized out>) at mailbox.cpp:82
>> #8  0x0000000000c1a9a6 in zmq::socket_base_t::process_commands
>> (this=this at entry=0x7f18be60e100,
>>     timeout_=<optimized out>, throttle_=throttle_ at entry=false) at
>> socket_base.cpp:884 #9  0x0000000000c1b11a in zmq::socket_base_t::recv
>> (this=this at entry=0x7f18be60e100,
>>     msg_=msg_ at entry=0x7f18b9ffdf40, flags_=flags_ at entry=0) at
>> socket_base.cpp:818 #10 0x0000000000c09d99 in s_recvmsg
>> (s_=s_ at entry=0x7f18be60e100, msg_=msg_ at entry=0x7f18b9ffdf40,
>>     flags_=flags_ at entry=0) at zmq.cpp:460
>
> Are you using multiple threads on the same socket? in case don't do it.
> If not, I have the same issue as posted already in the list..
>
> _______________________________________________
> zeromq-dev mailing list
> zeromq-dev at lists.zeromq.org
> http://lists.zeromq.org/mailman/listinfo/zeromq-dev