amafarinha
commented
5 years ago I'm facing the same issue, the output of gdb's backtrace is:
0 0x000000000052245c in el::LogMessage::level() const ()
1 0x000000000051b9cd in el::base::DefaultLogDispatchCallback::dispatch(std::__cxx11::basic_string<char, std::char_traits, std::allocator >&&) ()
2 0x000000000051b799 in el::base::DefaultLogDispatchCallback::handle(el::LogDispatchData const*) ()
It also happens to me that it prints some DEBUG and TRACE logs although they are disable in the configs... as in #700
justinasvd
fengcanyue
Hi, I'm getting segmentation fault when running multiple loggers via multiple threads. This is similar to issue #580 I have used the new version v.9.96.7 which supposedly fixes this issue since v 9.96.4 (stated in #580 ) I have also implemented the
define ELPP_NO_GLOBAL_LOCK
but the segmentation fault still persist randomly
My Configuration implements
define ELPP_THREAD_SAFE
I've also tried implementing the ELPP_THREAD_SAFE option via the compiler as in issue #314 but it doesn't fix the problem
On Failure gdb show this ... Running v.9.96.7
Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7fffa7fff700 (LWP 2379)] el::base::DefaultLogDispatchCallback::dispatch(std::string&&) (this=0x0, this@entry=0xaffe40, logLine=logLine@entry=<unknown type in /root/Documents/MagicEye/SUSI/testLib2/testLib2, CU 0x24caec, DIE 0x2f7c28>) at easylogging++.cc:2240 2240 if (m_data->logMessage()->logger()->m_typedConfigurations->toStandardOutput(m_data->logMessage()->level())) {
BackTrace
0 el::base::DefaultLogDispatchCallback::dispatch(std::string&&) (this=0x0, this@entry=0xaffe40,
1 0x0000000000529946 in el::base::DefaultLogDispatchCallback::handle (this=0xaffe40,
================================================= It always fail at the same location.
Previously with version 9.95.3, it will fail at the same location as in #580 Program terminated with signal SIGSEGV, Segmentation fault.
0 el::base::DefaultLogDispatchCallback::dispatch(std::string&&) (this=0xfb40f0,
Any Ideas about sorting the issue out? Thanks in advance ...
Regards Rod