Core dump with fluid distributed training
Created by: Yancey1989
I'm tring to run word2vec with distributed training, but it was core dump with VLOG_v=3
> GLOG_v=3 PSERVERS=127.0.0.1:6479 SERVER_ENDPOINT=127.0.0.1:6479 TRAINING_ROLE=PSERVER python test_word2vec.py
*** Aborted at 1515403120 (unix time) try "date -d @1515403120" if you are using GNU date ***
PC: @ 0x0 (unknown)
*** SIGSEGV (@0x0) received by PID 29809 (TID 0x7f0a93d05700) from PID 0; stack trace: ***
@ 0x7f0a938e2390 (unknown)
@ 0x7f0a4a7c8f52 _ZNK6paddle9framework12OperatorBase13DebugStringExB5cxx11EPKNS0_5ScopeE
@ 0x7f0a4a30d962 paddle::framework::Executor::Run()
@ 0x7f0a4a27a243 _ZZN8pybind1112cpp_function10initializeIZNS0_C4IvN6paddle9framework8ExecutorEIRKNS4_11ProgramDescEPNS4_5ScopeEibbEINS_4nameENS_9is_methodENS_7siblingEEEEMT0_FT_DpT1_EDpRKT2_EUlPS5_S8_SA_ibbE_vISO_S8_SA_ibbEISB_SC_SD_EEEvOSF_PFSE_SH_ESN_ENUlRNS_6detail13function_callEE1_4_FUNESV_
@ 0x7f0a4a277e94 pybind11::cpp_function::dispatcher()
@ 0x4cada2 PyEval_EvalFrameEx
@ 0x4c2765 PyEval_EvalCodeEx
@ 0x4ca099 PyEval_EvalFrameEx
@ 0x4c2765 PyEval_EvalCodeEx
@ 0x4c2509 PyEval_EvalCode
@ 0x4f1def (unknown)
@ 0x4ec652 PyRun_FileExFlags
@ 0x4eae31 PyRun_SimpleFileExFlags
@ 0x49e14a Py_Main
@ 0x7f0a93527830 __libc_start_main
@ 0x49d9d9 _start
@ 0x0 (unknown)