PaddleNLP pipeline server init error

欢迎您反馈PaddleNLP使用问题，非常感谢您对PaddleNLP的贡献！在留下您的问题时，辛苦您同步提供如下信息：

版本、环境信息 1）PaddleNLP和PaddlePaddle版本：请提供您的PaddleNLP和PaddlePaddle版本号，例如PaddleNLP 2.2.6 PaddlePaddle2.2.2 2）系统环境：请您描述系统类型，Windows，python3.9
复现信息：如为报错，请给出复现环境、复现步骤最后Paddle Serving部署时出错pipeline server init error Traceback (most recent call last): File "C:\Users\666\anaconda3\lib\site-packages\paddle_serving_server\pipeline\error_catch.py", line 97, in wrapper res = func(*args, **kw) File "C:\Users\666\anaconda3\lib\site-packages\paddle_serving_server\pipeline\error_catch.py", line 163, in wrapper result = function(*args, **kwargs) File "C:\Users\666\anaconda3\lib\site-packages\paddle_serving_server\pipeline\pipeline_server.py", line 51, in init_helper self._dag_executor = dag.DAGExecutor(response_op, dag_conf, worker_idx) File "C:\Users\666\anaconda3\lib\site-packages\paddle_serving_server\pipeline\dag.py", line 85, in init self._tracer = PerformanceTracer( File "C:\Users\666\anaconda3\lib\site-packages\paddle_serving_server\pipeline\profiler.py", line 45, in init self._data_buffer = multiprocessing.Manager().Queue() File "C:\Users\666\anaconda3\lib\multiprocessing\context.py", line 57, in Manager m.start() File "C:\Users\666\anaconda3\lib\multiprocessing\managers.py", line 554, in start self._process.start() File "C:\Users\666\anaconda3\lib\multiprocessing\process.py", line 121, in start self._popen = self._Popen(self) File "C:\Users\666\anaconda3\lib\multiprocessing\context.py", line 327, in _Popen return Popen(process_obj) File "C:\Users\666\anaconda3\lib\multiprocessing\popen_spawn_win32.py", line 45, in init prep_data = spawn.get_preparation_data(process_obj._name) File "C:\Users\666\anaconda3\lib\multiprocessing\spawn.py", line 154, in get_preparation_data _check_not_importing_main() File "C:\Users\666\anaconda3\lib\multiprocessing\spawn.py", line 134, in _check_not_importing_main raise RuntimeError(''' RuntimeError: An attempt has been made to start a new process before the current process has finished its bootstrapping phase.
```
  This probably means that you are not using fork to start your
  child processes and you have forgotten to use the proper idiom
  in the main module:

      if __name__ == '__main__':
          freeze_support()
          ...

  The "freeze_support()" line can be omitted if the program
  is not going to be frozen to produce an executable.
```

Classname: PipelineServicer.init..init_helper FunctionName: init_helper

May 16 '22 07:05 wenjin11

您好，请问是运行什么程序导致的错误？ @wenjin11

May 16 '22 08:05 w5688414

您好，请问是运行什么程序导致的错误？ @wenjin11

部署服务端运行web_service.py

May 16 '22 08:05 wenjin11

您好，请问是运行什么程序导致的错误？ @wenjin11

faq_system项目中的

May 16 '22 08:05 wenjin11

请问您在哪个群，可以艾特一下，咱们沟通一下？

May 16 '22 09:05 w5688414

交流三群我不知道您的群名称或Q名这是我的2541352748

May 16 '22 09:05 wenjin11

这是原生Windows系统上运行吗？

May 16 '22 10:05 TeslaZhao

serving目前没有适配windows，可以使用linux或者docker环境

May 16 '22 10:05 w5688414

@w5688414 同样的项目，macos 下也有报错，报错信息如下，麻烦请问该如何解决呢，谢谢

❯ python web_service.py
config use_encryption model here None
config use_encryption model here None
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/spawn.py", line 116, in spawn_main
    exitcode = _main(fd, parent_sentinel)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/spawn.py", line 125, in _main
    prepare(preparation_data)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/spawn.py", line 236, in prepare
    _fixup_main_from_path(data['init_main_from_path'])
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/spawn.py", line 287, in _fixup_main_from_path
    main_content = runpy.run_path(main_path,
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/runpy.py", line 265, in run_path
    return _run_module_code(code, init_globals, run_name,
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/runpy.py", line 97, in _run_module_code
    _run_code(code, mod_globals, init_globals,
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/Users/luxu/Documents/faq_system/deploy/python/web_service.py", line 82, in <module>
    ernie_service.run_service()
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/web_service.py", line 69, in run_service
    self._server.run_server()
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/pipeline/pipeline_server.py", line 314, in run_server
    PipelineServicer(self._name, self._response_op, self._conf),
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/pipeline/pipeline_server.py", line 57, in __init__
    raise CustomException(CustomExceptionCode.INIT_ERROR, "pipeline server init error")
paddle_serving_server.pipeline.error_catch.CustomException: 
        exception_code: 3003
        exception_type: INIT_ERROR
        error_msg: pipeline server init error
        is_send_to_user: False
Traceback (most recent call last):
  File "web_service.py", line 82, in <module>
    ernie_service.run_service()
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/web_service.py", line 69, in run_service
    self._server.run_server()
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/pipeline/pipeline_server.py", line 314, in run_server
    PipelineServicer(self._name, self._response_op, self._conf),
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/pipeline/pipeline_server.py", line 57, in __init__
    raise CustomException(CustomExceptionCode.INIT_ERROR, "pipeline server init error")
paddle_serving_server.pipeline.error_catch.CustomException: 
        exception_code: 3003
        exception_type: INIT_ERROR
        error_msg: pipeline server init error
        is_send_to_user: False

Jun 29 '22 02:06 luxu1220

将 is_thread_op 改为 true 后

# op资源类型, True, 为线程模型；False，为进程模型
  is_thread_op: True

报错变为

❯ python web_service.py
config use_encryption model here None
[DAG] Succ init
E0629 11:19:20.553887 302106112 analysis_config.cc:95] Please compile with gpu to EnableGpu()
--- Running analysis [ir_graph_build_pass]
--- Running analysis [ir_graph_clean_pass]
--- Running analysis [ir_analysis_pass]
--- Running IR pass [simplify_with_basic_ops_pass]
--- Running IR pass [layer_norm_fuse_pass]
---    Fused 0 subgraphs into layer_norm op.
--- Running IR pass [attention_lstm_fuse_pass]
--- Running IR pass [seqconv_eltadd_relu_fuse_pass]
--- Running IR pass [seqpool_cvm_concat_fuse_pass]
--- Running IR pass [mul_lstm_fuse_pass]
--- Running IR pass [fc_gru_fuse_pass]
---    fused 0 pairs of fc gru patterns
--- Running IR pass [mul_gru_fuse_pass]
--- Running IR pass [seq_concat_fc_fuse_pass]
--- Running IR pass [gpu_cpu_squeeze2_matmul_fuse_pass]
--- Running IR pass [gpu_cpu_reshape2_matmul_fuse_pass]
--- Running IR pass [gpu_cpu_flatten2_matmul_fuse_pass]
--- Running IR pass [matmul_v2_scale_fuse_pass]
--- Running IR pass [gpu_cpu_map_matmul_v2_to_mul_pass]
I0629 11:19:21.893158 302106112 fuse_pass_base.cc:57] ---  detected 74 subgraphs
--- Running IR pass [gpu_cpu_map_matmul_v2_to_matmul_pass]
I0629 11:19:21.896107 302106112 fuse_pass_base.cc:57] ---  detected 12 subgraphs
--- Running IR pass [matmul_scale_fuse_pass]
--- Running IR pass [gpu_cpu_map_matmul_to_mul_pass]
--- Running IR pass [fc_fuse_pass]
I0629 11:19:22.389168 302106112 fuse_pass_base.cc:57] ---  detected 74 subgraphs
--- Running IR pass [repeated_fc_relu_fuse_pass]
--- Running IR pass [squared_mat_sub_fuse_pass]
--- Running IR pass [conv_bn_fuse_pass]
--- Running IR pass [conv_eltwiseadd_bn_fuse_pass]
--- Running IR pass [conv_transpose_bn_fuse_pass]
--- Running IR pass [is_test_pass]
--- Running IR pass [runtime_context_cache_pass]
--- Running analysis [ir_params_sync_among_devices_pass]
--- Running analysis [adjust_cudnn_workspace_size_pass]
--- Running analysis [inference_op_replace_pass]
--- Running analysis [memory_optimize_pass]
I0629 11:19:22.469272 302106112 memory_optimize_pass.cc:216] Cluster name : token_type_ids  size: 8
I0629 11:19:22.469298 302106112 memory_optimize_pass.cc:216] Cluster name : matmul_33.tmp_0  size: 48
I0629 11:19:22.469305 302106112 memory_optimize_pass.cc:216] Cluster name : relu_28.tmp_0  size: 12288
I0629 11:19:22.469309 302106112 memory_optimize_pass.cc:216] Cluster name : layer_norm_93.tmp_2  size: 3072
I0629 11:19:22.469314 302106112 memory_optimize_pass.cc:216] Cluster name : transpose_140.tmp_0  size: 3072
I0629 11:19:22.469318 302106112 memory_optimize_pass.cc:216] Cluster name : layer_norm_89.tmp_2  size: 3072
I0629 11:19:22.469322 302106112 memory_optimize_pass.cc:216] Cluster name : unsqueeze2_2.tmp_0  size: 4
I0629 11:19:22.469326 302106112 memory_optimize_pass.cc:216] Cluster name : layer_norm_78.tmp_2  size: 3072
--- Running analysis [ir_graph_to_program_pass]
I0629 11:19:22.614665 302106112 analysis_predictor.cc:1007] ======= optimize end =======
I0629 11:19:22.615101 302106112 naive_executor.cc:102] ---  skip [feed], feed -> token_type_ids
I0629 11:19:22.616220 302106112 naive_executor.cc:102] ---  skip [feed], feed -> input_ids
I0629 11:19:22.621551 302106112 naive_executor.cc:102] ---  skip [elementwise_div_2], fetch -> fetch
[PipelineServicer] succ init
Traceback (most recent call last):
  File "web_service.py", line 82, in <module>
    ernie_service.run_service()
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/web_service.py", line 69, in run_service
    self._server.run_server()
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/pipeline/pipeline_server.py", line 318, in run_server
    self._run_grpc_gateway(
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/pipeline/pipeline_server.py", line 148, in _run_grpc_gateway
    self._proxy_server.start()
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/process.py", line 121, in start
    self._popen = self._Popen(self)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/context.py", line 224, in _Popen
    return _default_context.get_context().Process._Popen(process_obj)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/context.py", line 284, in _Popen
    return Popen(process_obj)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/popen_spawn_posix.py", line 32, in __init__
    super().__init__(process_obj)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/popen_fork.py", line 19, in __init__
    self._launch(process_obj)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/popen_spawn_posix.py", line 47, in _launch
    reduction.dump(process_obj, fp)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/reduction.py", line 60, in dump
    ForkingPickler(file, protocol).dump(obj)
TypeError: cannot pickle '_thread.lock' object

Jun 29 '22 03:06 luxu1220

我也碰到了同样的问题，最后修改config.yml中模型文件夹的名称为正确名称就可以了。

Aug 02 '22 09:08 zhaobinchen

我也遇到问题，请问后来是如何解决的，谢谢

Oct 27 '22 05:10 JaylonXiwei

将 is_thread_op 改为 true 后

# op资源类型, True, 为线程模型；False，为进程模型
  is_thread_op: True

报错变为

❯ python web_service.py
config use_encryption model here None
[DAG] Succ init
E0629 11:19:20.553887 302106112 analysis_config.cc:95] Please compile with gpu to EnableGpu()
--- Running analysis [ir_graph_build_pass]
--- Running analysis [ir_graph_clean_pass]
--- Running analysis [ir_analysis_pass]
--- Running IR pass [simplify_with_basic_ops_pass]
--- Running IR pass [layer_norm_fuse_pass]
---    Fused 0 subgraphs into layer_norm op.
--- Running IR pass [attention_lstm_fuse_pass]
--- Running IR pass [seqconv_eltadd_relu_fuse_pass]
--- Running IR pass [seqpool_cvm_concat_fuse_pass]
--- Running IR pass [mul_lstm_fuse_pass]
--- Running IR pass [fc_gru_fuse_pass]
---    fused 0 pairs of fc gru patterns
--- Running IR pass [mul_gru_fuse_pass]
--- Running IR pass [seq_concat_fc_fuse_pass]
--- Running IR pass [gpu_cpu_squeeze2_matmul_fuse_pass]
--- Running IR pass [gpu_cpu_reshape2_matmul_fuse_pass]
--- Running IR pass [gpu_cpu_flatten2_matmul_fuse_pass]
--- Running IR pass [matmul_v2_scale_fuse_pass]
--- Running IR pass [gpu_cpu_map_matmul_v2_to_mul_pass]
I0629 11:19:21.893158 302106112 fuse_pass_base.cc:57] ---  detected 74 subgraphs
--- Running IR pass [gpu_cpu_map_matmul_v2_to_matmul_pass]
I0629 11:19:21.896107 302106112 fuse_pass_base.cc:57] ---  detected 12 subgraphs
--- Running IR pass [matmul_scale_fuse_pass]
--- Running IR pass [gpu_cpu_map_matmul_to_mul_pass]
--- Running IR pass [fc_fuse_pass]
I0629 11:19:22.389168 302106112 fuse_pass_base.cc:57] ---  detected 74 subgraphs
--- Running IR pass [repeated_fc_relu_fuse_pass]
--- Running IR pass [squared_mat_sub_fuse_pass]
--- Running IR pass [conv_bn_fuse_pass]
--- Running IR pass [conv_eltwiseadd_bn_fuse_pass]
--- Running IR pass [conv_transpose_bn_fuse_pass]
--- Running IR pass [is_test_pass]
--- Running IR pass [runtime_context_cache_pass]
--- Running analysis [ir_params_sync_among_devices_pass]
--- Running analysis [adjust_cudnn_workspace_size_pass]
--- Running analysis [inference_op_replace_pass]
--- Running analysis [memory_optimize_pass]
I0629 11:19:22.469272 302106112 memory_optimize_pass.cc:216] Cluster name : token_type_ids  size: 8
I0629 11:19:22.469298 302106112 memory_optimize_pass.cc:216] Cluster name : matmul_33.tmp_0  size: 48
I0629 11:19:22.469305 302106112 memory_optimize_pass.cc:216] Cluster name : relu_28.tmp_0  size: 12288
I0629 11:19:22.469309 302106112 memory_optimize_pass.cc:216] Cluster name : layer_norm_93.tmp_2  size: 3072
I0629 11:19:22.469314 302106112 memory_optimize_pass.cc:216] Cluster name : transpose_140.tmp_0  size: 3072
I0629 11:19:22.469318 302106112 memory_optimize_pass.cc:216] Cluster name : layer_norm_89.tmp_2  size: 3072
I0629 11:19:22.469322 302106112 memory_optimize_pass.cc:216] Cluster name : unsqueeze2_2.tmp_0  size: 4
I0629 11:19:22.469326 302106112 memory_optimize_pass.cc:216] Cluster name : layer_norm_78.tmp_2  size: 3072
--- Running analysis [ir_graph_to_program_pass]
I0629 11:19:22.614665 302106112 analysis_predictor.cc:1007] ======= optimize end =======
I0629 11:19:22.615101 302106112 naive_executor.cc:102] ---  skip [feed], feed -> token_type_ids
I0629 11:19:22.616220 302106112 naive_executor.cc:102] ---  skip [feed], feed -> input_ids
I0629 11:19:22.621551 302106112 naive_executor.cc:102] ---  skip [elementwise_div_2], fetch -> fetch
[PipelineServicer] succ init
Traceback (most recent call last):
  File "web_service.py", line 82, in <module>
    ernie_service.run_service()
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/web_service.py", line 69, in run_service
    self._server.run_server()
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/pipeline/pipeline_server.py", line 318, in run_server
    self._run_grpc_gateway(
  File "/Users/luxu/.pyenv/versions/paddle/lib/python3.8/site-packages/paddle_serving_server/pipeline/pipeline_server.py", line 148, in _run_grpc_gateway
    self._proxy_server.start()
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/process.py", line 121, in start
    self._popen = self._Popen(self)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/context.py", line 224, in _Popen
    return _default_context.get_context().Process._Popen(process_obj)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/context.py", line 284, in _Popen
    return Popen(process_obj)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/popen_spawn_posix.py", line 32, in __init__
    super().__init__(process_obj)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/popen_fork.py", line 19, in __init__
    self._launch(process_obj)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/popen_spawn_posix.py", line 47, in _launch
    reduction.dump(process_obj, fp)
  File "/Users/luxu/.pyenv/versions/3.8.12/lib/python3.8/multiprocessing/reduction.py", line 60, in dump
    ForkingPickler(file, protocol).dump(obj)
TypeError: cannot pickle '_thread.lock' object

mac m1 同样的问题

Oct 28 '22 07:10 dougzhi

This issue is stale because it has been open for 60 days with no activity. 当前issue 60天内无活动，被标记为stale。

Jan 06 '23 05:01 github-actions[bot]

This issue was closed because it has been inactive for 14 days since being marked as stale. 当前issue 被标记为stale已有14天，即将关闭。

Jan 21 '23 00:01 github-actions[bot]

PaddleNLP PaddleNLP copied to clipboard

pipeline server init error

PaddleNLP
PaddleNLP copied to clipboard