1、CPU：Intel(R) Xeon(R) Silver 4410Y

2、DCU：K100_AI * 4

3、镜像：image.sourcefind.cn:5000/dcu/admin/base/vllm:0.9.2-ubuntu22.04-dtk25.04.1-rc5-rocblas101839-0811-das1.6-py3.10-20250913-rc1

4、启动命令：

vllm serve /mnt/model/Qwen3-235B-A22B-GPTQ-Int4/   --tensor-parallel-size 4 --served-model-name Qwen3-235B-A22B-GPTQ-Int4  --gpu-memory-utilization 0.85 -tp 4 --enable-reasoning

5、报错信息：

(VllmWorker rank=0 pid=3855) WARNING 09-19 10:18:55 [config.py:444] MoE DP setup unable to determine quantization scheme or unsupported quantization type. This model will not run with DP enabled.
(VllmWorker rank=2 pid=3857) INFO 09-19 10:18:55 [rocm.py:279] Using Flash Attention backend on V1 engine. (only supports block size 64)
(VllmWorker rank=3 pid=3858) INFO 09-19 10:18:55 [rocm.py:279] Using Flash Attention backend on V1 engine. (only supports block size 64)
(VllmWorker rank=2 pid=3857) WARNING 09-19 10:18:55 [config.py:444] MoE DP setup unable to determine quantization scheme or unsupported quantization type. This model will not run with DP enabled.
(VllmWorker rank=3 pid=3858) WARNING 09-19 10:18:55 [config.py:444] MoE DP setup unable to determine quantization scheme or unsupported quantization type. This model will not run with DP enabled.
(VllmWorker rank=1 pid=3856) INFO 09-19 10:18:55 [rocm.py:279] Using Flash Attention backend on V1 engine. (only supports block size 64)
(VllmWorker rank=1 pid=3856) WARNING 09-19 10:18:55 [config.py:444] MoE DP setup unable to determine quantization scheme or unsupported quantization type. This model will not run with DP enabled.
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487] WorkerProc failed to start.
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487] Traceback (most recent call last):
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/executor/multiproc_executor.py", line 461, in worker_main
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     worker = WorkerProc(*args, **kwargs)
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/executor/multiproc_executor.py", line 358, in __init__
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     self.worker.load_model()
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/worker/gpu_worker.py", line 192, in load_model
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     self.model_runner.load_model()
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 1794, in load_model
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     self.model = model_loader.load_model(
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/model_loader/base_loader.py", line 38, in load_model
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     model = initialize_model(vllm_config=vllm_config,
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/model_loader/utils.py", line 67, in initialize_model
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     return model_class(vllm_config=vllm_config, prefix=prefix)
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/models/qwen3_moe.py", line 575, in __init__
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     self.model = Qwen3MoeModel(vllm_config=vllm_config,
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/compilation/decorators.py", line 154, in __init__
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     old_init(self, vllm_config=vllm_config, prefix=prefix, **kwargs)
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/models/qwen3_moe.py", line 353, in __init__
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     self.start_layer, self.end_layer, self.layers = make_layers(
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/models/utils.py", line 639, in make_layers
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     [PPMissingLayer() for _ in range(start_layer)] + [
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/models/utils.py", line 640, in <listcomp>
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     maybe_offload_to_cpu(layer_fn(prefix=f"{prefix}.{idx}"))
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/models/qwen3_moe.py", line 355, in <lambda>
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     lambda prefix: Qwen3MoeDecoderLayer(config=config,
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/models/qwen3_moe.py", line 290, in __init__
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     self.mlp = Qwen3MoeSparseMoeBlock(config=config,
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/models/qwen3_moe.py", line 119, in __init__
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     self.experts = FusedMoE(num_experts=config.num_experts,
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]   File "/usr/local/lib/python3.10/dist-packages/vllm/model_executor/layers/fused_moe/layer.py", line 781, in __init__
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487]     assert quant_method is not None
(VllmWorker rank=0 pid=3855) ERROR 09-19 10:18:55 [multiproc_executor.py:487] AssertionError
I0919 10:18:56.467299  3855 ProcessGroupNCCL.cpp:1263] [PG ID 0 PG GUID 0 Rank 0] ProcessGroupNCCL destructor entered.
W0919 10:18:56.467343  3855 ProcessGroupNCCL.cpp:1267] Warning: WARNING: process group has NOT been destroyed before we destruct ProcessGroupNCCL. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL operations have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4 (function operator())
I0919 10:18:56.467356  3855 ProcessGroupNCCL.cpp:1247] [PG ID 0 PG GUID 0 Rank 0] Launching ProcessGroupNCCL abort asynchrounously.
I0919 10:18:56.467491  3855 ProcessGroupNCCL.cpp:1133] [PG ID 0 PG GUID 0 Rank 0] future is successfully executed for: ProcessGroup abort
I0919 10:18:56.467496  3855 ProcessGroupNCCL.cpp:1254] [PG ID 0 PG GUID 0 Rank 0] ProcessGroupNCCL aborts successfully.
I0919 10:18:56.467522  3855 ProcessGroupNCCL.cpp:1284] [PG ID 0 PG GUID 0 Rank 0] ProcessGroupNCCL watchdog thread joined.
I0919 10:18:56.467543  3855 ProcessGroupNCCL.cpp:1288] [PG ID 0 PG GUID 0 Rank 0] ProcessGroupNCCL heart beat monitor thread joined.
I0919 10:18:56.491492  3857 ProcessGroupNCCL.cpp:1263] [PG ID 0 PG GUID 0 Rank 2] ProcessGroupNCCL destructor entered.
I0919 10:18:56.491536  3857 ProcessGroupNCCL.cpp:1247] [PG ID 0 PG GUID 0 Rank 2] Launching ProcessGroupNCCL abort asynchrounously.
I0919 10:18:56.491669  3857 ProcessGroupNCCL.cpp:1133] [PG ID 0 PG GUID 0 Rank 2] future is successfully executed for: ProcessGroup abort
I0919 10:18:56.491676  3857 ProcessGroupNCCL.cpp:1254] [PG ID 0 PG GUID 0 Rank 2] ProcessGroupNCCL aborts successfully.
I0919 10:18:56.491703  3857 ProcessGroupNCCL.cpp:1284] [PG ID 0 PG GUID 0 Rank 2] ProcessGroupNCCL watchdog thread joined.
I0919 10:18:56.491730  3857 ProcessGroupNCCL.cpp:1288] [PG ID 0 PG GUID 0 Rank 2] ProcessGroupNCCL heart beat monitor thread joined.
I0919 10:18:56.507534  3858 ProcessGroupNCCL.cpp:1263] [PG ID 0 PG GUID 0 Rank 3] ProcessGroupNCCL destructor entered.
I0919 10:18:56.507573  3858 ProcessGroupNCCL.cpp:1247] [PG ID 0 PG GUID 0 Rank 3] Launching ProcessGroupNCCL abort asynchrounously.
I0919 10:18:56.507689  3858 ProcessGroupNCCL.cpp:1133] [PG ID 0 PG GUID 0 Rank 3] future is successfully executed for: ProcessGroup abort
I0919 10:18:56.507694  3858 ProcessGroupNCCL.cpp:1254] [PG ID 0 PG GUID 0 Rank 3] ProcessGroupNCCL aborts successfully.
I0919 10:18:56.507725  3858 ProcessGroupNCCL.cpp:1284] [PG ID 0 PG GUID 0 Rank 3] ProcessGroupNCCL watchdog thread joined.
I0919 10:18:56.507755  3858 ProcessGroupNCCL.cpp:1288] [PG ID 0 PG GUID 0 Rank 3] ProcessGroupNCCL heart beat monitor thread joined.
I0919 10:18:56.698796  3856 ProcessGroupNCCL.cpp:1263] [PG ID 0 PG GUID 0 Rank 1] ProcessGroupNCCL destructor entered.
I0919 10:18:56.698851  3856 ProcessGroupNCCL.cpp:1247] [PG ID 0 PG GUID 0 Rank 1] Launching ProcessGroupNCCL abort asynchrounously.
I0919 10:18:56.699033  3856 ProcessGroupNCCL.cpp:1133] [PG ID 0 PG GUID 0 Rank 1] future is successfully executed for: ProcessGroup abort
I0919 10:18:56.699038  3856 ProcessGroupNCCL.cpp:1254] [PG ID 0 PG GUID 0 Rank 1] ProcessGroupNCCL aborts successfully.
I0919 10:18:56.699066  3856 ProcessGroupNCCL.cpp:1284] [PG ID 0 PG GUID 0 Rank 1] ProcessGroupNCCL watchdog thread joined.
I0919 10:18:56.699100  3856 ProcessGroupNCCL.cpp:1288] [PG ID 0 PG GUID 0 Rank 1] ProcessGroupNCCL heart beat monitor thread joined.
ERROR 09-19 10:18:59 [core.py:589] EngineCore failed to start.
ERROR 09-19 10:18:59 [core.py:589] Traceback (most recent call last):
ERROR 09-19 10:18:59 [core.py:589]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core.py", line 580, in run_engine_core
ERROR 09-19 10:18:59 [core.py:589]     engine_core = EngineCoreProc(*args, **kwargs)
ERROR 09-19 10:18:59 [core.py:589]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core.py", line 407, in __init__
ERROR 09-19 10:18:59 [core.py:589]     super().__init__(vllm_config, executor_class, log_stats,
ERROR 09-19 10:18:59 [core.py:589]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core.py", line 77, in __init__
ERROR 09-19 10:18:59 [core.py:589]     self.model_executor = executor_class(vllm_config)
ERROR 09-19 10:18:59 [core.py:589]   File "/usr/local/lib/python3.10/dist-packages/vllm/executor/executor_base.py", line 53, in __init__
ERROR 09-19 10:18:59 [core.py:589]     self._init_executor()
ERROR 09-19 10:18:59 [core.py:589]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/executor/multiproc_executor.py", line 93, in _init_executor
ERROR 09-19 10:18:59 [core.py:589]     self.workers = WorkerProc.wait_for_ready(unready_workers)
ERROR 09-19 10:18:59 [core.py:589]   File "/usr/local/lib/python3.10/dist-packages/vllm/v1/executor/multiproc_executor.py", line 422, in wait_for_ready
ERROR 09-19 10:18:59 [core.py:589]     raise e from None
ERROR 09-19 10:18:59 [core.py:589] Exception: WorkerProc initialization failed due to an exception in a background process. See stack trace for root cause.
Process EngineCore_0:
Traceback (most recent call last):
  File "/usr/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap
    self.run()
  File "/usr/lib/python3.10/multiprocessing/process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core.py", line 593, in run_engine_core
    raise e
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core.py", line 580, in run_engine_core
    engine_core = EngineCoreProc(*args, **kwargs)
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core.py", line 407, in __init__
    super().__init__(vllm_config, executor_class, log_stats,
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core.py", line 77, in __init__
    self.model_executor = executor_class(vllm_config)
  File "/usr/local/lib/python3.10/dist-packages/vllm/executor/executor_base.py", line 53, in __init__
    self._init_executor()
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/executor/multiproc_executor.py", line 93, in _init_executor
    self.workers = WorkerProc.wait_for_ready(unready_workers)
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/executor/multiproc_executor.py", line 422, in wait_for_ready
    raise e from None
Exception: WorkerProc initialization failed due to an exception in a background process. See stack trace for root cause.
Traceback (most recent call last):
  File "/usr/local/bin/vllm", line 8, in <module>
    sys.exit(main())
  File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/cli/main.py", line 65, in main
    args.dispatch_function(args)
  File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/cli/serve.py", line 55, in cmd
    uvloop.run(run_server(args))
  File "/usr/local/lib/python3.10/dist-packages/uvloop/__init__.py", line 82, in run
    return loop.run_until_complete(wrapper())
  File "uvloop/loop.pyx", line 1518, in uvloop.loop.Loop.run_until_complete
  File "/usr/local/lib/python3.10/dist-packages/uvloop/__init__.py", line 61, in wrapper
    return await main
  File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 1431, in run_server
    await run_server_worker(listen_address, sock, args, **uvicorn_kwargs)
  File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 1451, in run_server_worker
    async with build_async_engine_client(args, client_config) as engine_client:
  File "/usr/lib/python3.10/contextlib.py", line 199, in __aenter__
    return await anext(self.gen)
  File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 158, in build_async_engine_client
    async with build_async_engine_client_from_engine_args(
  File "/usr/lib/python3.10/contextlib.py", line 199, in __aenter__
    return await anext(self.gen)
  File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 194, in build_async_engine_client_from_engine_args
    async_llm = AsyncLLM.from_vllm_config(
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/async_llm.py", line 162, in from_vllm_config
    return cls(
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/async_llm.py", line 124, in __init__
    self.engine_core = EngineCoreClient.make_async_mp_client(
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core_client.py", line 96, in make_async_mp_client
    return AsyncMPClient(*client_args)
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core_client.py", line 666, in __init__
    super().__init__(
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/core_client.py", line 403, in __init__
    with launch_core_engines(vllm_config, executor_class,
  File "/usr/lib/python3.10/contextlib.py", line 142, in __exit__
    next(self.gen)
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/utils.py", line 434, in launch_core_engines
    wait_for_engine_startup(
  File "/usr/local/lib/python3.10/dist-packages/vllm/v1/engine/utils.py", line 484, in wait_for_engine_startup
    raise RuntimeError("Engine core initialization failed. "
RuntimeError: Engine core initialization failed. See root cause above. Failed core proc(s): {}
root@aicloud247:/workspace# /usr/lib/python3.10/multiprocessing/resource_tracker.py:224: UserWarning: resource_tracker: There appear to be 1 leaked shared_memory objects to clean up at shutdown
  warnings.warn('resource_tracker: There appear to be %d '