[GitHub] [incubator-mxnet-ci] ChaiBapchya commented on pull request #23: [DO NOT MERGE] Test g3

[GitBox <gi...@apache.org>]

ChaiBapchya commented on pull request #23:
URL: https://github.com/apache/incubator-mxnet-ci/pull/23#issuecomment-618736828


   Created an AMI using the above mentioned scripts & following the steps as mentioned in the updated Readme.
   
   Launched a G3.8xl instance with the created AMI.
   ```
   sudo python3 ci/build.py --platform ubuntu_build_cuda /work/runtime_functions.sh build_ubuntu_gpu_cuda101_cudnn7
   sudo python3 ci/build.py --platform ubuntu_gpu_cu101 /work/runtime_functions.sh integrationtest_ubuntu_gpu_python
   ```
   Fail due to libnvidia-ml.so.1 not found
   ```
   2020-04-23 23:35:41,571 - root - INFO - Started container: fe0a2fbcd8eb
   + CI_CUDA_COMPUTE_CAPABILITIES='-gencode=arch=compute_52,code=sm_52 -gencode=arch=compute_70,code=sm_70'
   + CI_CMAKE_CUDA_ARCH='5.2 7.0'
   + set +x
   + export PYTHONPATH=./python/
   + PYTHONPATH=./python/
   + export MXNET_STORAGE_FALLBACK_LOG_VERBOSE=0
   + MXNET_STORAGE_FALLBACK_LOG_VERBOSE=0
   + export MXNET_SUBGRAPH_VERBOSE=0
   + MXNET_SUBGRAPH_VERBOSE=0
   + export DMLC_LOG_STACK_TRACE_DEPTH=10
   + DMLC_LOG_STACK_TRACE_DEPTH=10
   + pytest example/image-classification/test_score.py
   WARNING:root:Unable to import numpy/mxnet. Skipping conftest.
   ============================= test session starts ==============================
   platform linux -- Python 3.5.2, pytest-5.3.5, py-1.8.1, pluggy-0.13.1
   rootdir: /work/mxnet, inifile: pytest.ini
   plugins: timeout-1.3.4, xdist-1.31.0, forked-1.1.3, env-0.6.2, cov-2.8.1
   timeout: 1200.0s
   timeout method: signal
   timeout func_only: False
   collected 0 items / 1 error
   
   ==================================== ERRORS ====================================
   _________ ERROR collecting example/image-classification/test_score.py __________
   example/image-classification/test_score.py:25: in <module>
       import mxnet as mx
   python/mxnet/__init__.py:23: in <module>
       from .context import Context, current_context, cpu, gpu, cpu_pinned
   python/mxnet/context.py:23: in <module>
       from .base import classproperty, with_metaclass, _MXClassPropertyMetaClass
   python/mxnet/base.py:351: in <module>
       _LIB = _load_lib()
   python/mxnet/base.py:342: in _load_lib
       lib = ctypes.CDLL(lib_path[0], ctypes.RTLD_LOCAL)
   /usr/lib/python3.5/ctypes/__init__.py:347: in __init__
       self._handle = _dlopen(self._name, mode)
   E   OSError: libnvidia-ml.so.1: cannot open shared object file: No such file or directory
   !!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!
   =============================== 1 error in 0.13s ===============================
   2020-04-23 23:35:42,274 - root - INFO - Waiting for status of container fe0a2fbcd8eb for 600 s.
   2020-04-23 23:35:42,445 - root - INFO - Container exit status: {'StatusCode': 2, 'Error': None}
   2020-04-23 23:35:42,445 - root - ERROR - Container exited with an error 😞
   2020-04-23 23:35:42,445 - root - INFO - Executed command for reproduction:
   
   ci/build.py --platform ubuntu_gpu_cu101 /work/runtime_functions.sh integrationtest_ubuntu_gpu_python
   
   2020-04-23 23:35:42,446 - root - INFO - Stopping container: fe0a2fbcd8eb
   2020-04-23 23:35:42,447 - root - INFO - Removing container: fe0a2fbcd8eb
   2020-04-23 23:35:42,456 - root - CRITICAL - Execution of ['/work/runtime_functions.sh', 'integrationtest_ubuntu_gpu_python'] failed wit
   h status: 2
   ```


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org