Created by: yeyupiaoling

我使用的是百度深度学习GPU集群，以前使用没有问题的，今天新创建一个集群就报错，错误信息如下：

F0529 09:31:53.699651 39220 hl_cuda_device.cc:453] Check failed: cudaSuccess == cudaStat (0 vs. 38) Cuda Error: no CUDA-capable device is detected
*** Check failure stack trace: ***
    @     0x7f253560f18d  google::LogMessage::Fail()
    @     0x7f25356114d8  google::LogMessage::SendToLog()
    @     0x7f253560ec9b  google::LogMessage::Flush()
    @     0x7f25356123ae  google::LogMessageFatal::~LogMessageFatal()
    @     0x7f25355bce40  hl_specify_devices_start()
    @     0x7f25355bd04d  hl_start()
    @     0x7f25355496fe  paddle::initMain()
    @     0x7f25355f5761  initPaddle()
    @     0x7f25351690b7  _wrap_initPaddle
    @           0x4c45fa  PyEval_EvalFrameEx
    @           0x4c9d7f  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4de69e  (unknown)
    @           0x4b0c93  PyObject_Call
    @           0x4c6ef6  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4ca7df  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4ca7df  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4c24a9  PyEval_EvalCode
    @           0x4f19ef  (unknown)
    @           0x4ec372  PyRun_FileExFlags
    @           0x4eaaf1  PyRun_SimpleFileExFlags
    @           0x49e208  Py_Main
    @     0x7f2538a30830  __libc_start_main
    @           0x49da59  _start
    @              (nil)  (unknown)
Aborted (core dumped)

这个是PaddlePaddle的问题，还是服务器的问题呢？