本来想提交工单，但是提了八次都提示我提交失败，只能发帖求助o(╥﹏╥)o
我安装了docker版的vllm-ascend，启动大模型时报错，错误日志如下
驱动版本为25.0.rc1.1，固件版本为7.7.0.1.231, 均安装正常，系统为欧拉22.03，docker版本为24.0.8,安装的镜像为quay.io/ascend/vllm-ascend:v0.7.3-dev-openeuler
创建容器命令为
docker run -itd \
--name vllm-ascend-env \
--device /dev/davinci0 \
--device /dev/davinci1 \
--device /dev/davinci_manager \
--device /dev/devmm_svm \
--device /dev/hisi_hdc \
-v /usr/local/dcmi:/usr/local/dcmi \
-v /usr/local/bin/npu-smi:/usr/local/bin/npu-smi \
-v /usr/local/Ascend/driver/lib64/:/usr/local/Ascend/driver/lib64/ \
-v /usr/local/Ascend/driver/version.info:/usr/local/Ascend/driver/version.info \
-v /etc/ascend_install.info:/etc/ascend_install.info \
-v /home/models:/home/models \
-it quay.io/ascend/vllm-ascend:v0.7.3-dev-openeuler bash
启动大模型时报错 ，日志如下：
[root@09be6b1eee4f models]# vllm serve /home/models/Qwen2.5-7B --dtype auto --port 8000
INFO 06-24 16:50:05 __init__.py:30] Available plugins for group vllm.platform_plugins:
INFO 06-24 16:50:05 __init__.py:32] name=ascend, value=vllm_ascend:register
INFO 06-24 16:50:05 __init__.py:34] all available plugins for group vllm.platform_plugins will be loaded.
INFO 06-24 16:50:05 __init__.py:36] set environment variable VLLM_PLUGINS to control which plugins to load.
INFO 06-24 16:50:05 __init__.py:44] plugin ascend loaded.
INFO 06-24 16:50:05 __init__.py:198] Platform plugin ascend is activated
WARNING:root:Warning: Failed to register custom ops, all custom ops will be disabled
INFO 06-24 16:50:05 __init__.py:30] Available plugins for group vllm.general_plugins:
INFO 06-24 16:50:05 __init__.py:32] name=ascend_enhanced_model, value=vllm_ascend:register_model
INFO 06-24 16:50:05 __init__.py:34] all available plugins for group vllm.general_plugins will be loaded.
INFO 06-24 16:50:05 __init__.py:36] set environment variable VLLM_PLUGINS to control which plugins to load.
INFO 06-24 16:50:05 __init__.py:44] plugin ascend_enhanced_model loaded.
WARNING 06-24 16:50:05 registry.py:351] Model architecture Qwen2VLForConditionalGeneration is already registered, and will be overwritten by the new model class vllm_ascend.models.qwen2_vl:AscendQwen2VLForConditionalGeneration.
WARNING 06-24 16:50:05 registry.py:351] Model architecture Qwen2_5_VLForConditionalGeneration is already registered, and will be overwritten by the new model class vllm_ascend.models.qwen2_5_vl:AscendQwen2_5_VLForConditionalGeneration.
WARNING 06-24 16:50:05 registry.py:351] Model architecture DeepseekV2ForCausalLM is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_v2:CustomDeepseekV2ForCausalLM.
WARNING 06-24 16:50:05 registry.py:351] Model architecture DeepseekV3ForCausalLM is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_v2:CustomDeepseekV3ForCausalLM.
WARNING 06-24 16:50:05 registry.py:351] Model architecture DeepSeekMTPModel is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_mtp:CustomDeepSeekMTP.
INFO 06-24 16:50:05 importing.py:16] Triton not installed or not compatible; certain GPU-related functions will not be available.
INFO 06-24 16:50:05 api_server.py:912] vLLM API server version 0.7.3
INFO 06-24 16:50:05 api_server.py:913] args: Namespace(subparser='serve', model_tag='/home/models/Qwen2.5-7B', config='', host=None, port=8000, uvicorn_log_level='info', allow_credentials=False, allowed_origins=['*'], allowed_methods=['*'], allowed_headers=['*'], api_key=None, lora_modules=None, prompt_adapters=None, chat_template=None, chat_template_content_format='auto', response_role='assistant', ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_cert_reqs=0, root_path=None, middleware=[], return_tokens_as_token_ids=False, disable_frontend_multiprocessing=False, enable_request_id_headers=False, enable_auto_tool_choice=False, enable_reasoning=False, reasoning_parser=None, tool_call_parser=None, tool_parser_plugin='', model='/home/models/Qwen2.5-7B', task='auto', tokenizer=None, skip_tokenizer_init=False, revision=None, code_revision=None, tokenizer_revision=None, tokenizer_mode='auto', trust_remote_code=False, allowed_local_media_path=None, download_dir=None, load_format='auto', config_format=<ConfigFormat.AUTO: 'auto'>, dtype='auto', kv_cache_dtype='auto', max_model_len=None, guided_decoding_backend='xgrammar', logits_processor_pattern=None, model_impl='auto', distributed_executor_backend=None, pipeline_parallel_size=1, tensor_parallel_size=1, max_parallel_loading_workers=None, ray_workers_use_nsight=False, block_size=None, enable_prefix_caching=None, disable_sliding_window=False, use_v2_block_manager=True, num_lookahead_slots=0, seed=0, swap_space=4, cpu_offload_gb=0, gpu_memory_utilization=0.9, num_gpu_blocks_override=None, max_num_batched_tokens=None, max_num_partial_prefills=1, max_long_partial_prefills=1, long_prefill_token_threshold=0, max_num_seqs=None, max_logprobs=20, disable_log_stats=False, quantization=None, rope_scaling=None, rope_theta=None, hf_overrides=None, enforce_eager=False, max_seq_len_to_capture=8192, disable_custom_all_reduce=False, tokenizer_pool_size=0, tokenizer_pool_type='ray', tokenizer_pool_extra_config=None, limit_mm_per_prompt=None, mm_processor_kwargs=None, disable_mm_preprocessor_cache=False, enable_lora=False, enable_lora_bias=False, max_loras=1, max_lora_rank=16, lora_extra_vocab_size=256, lora_dtype='auto', long_lora_scaling_factors=None, max_cpu_loras=None, fully_sharded_loras=False, enable_prompt_adapter=False, max_prompt_adapters=1, max_prompt_adapter_token=0, device='auto', num_scheduler_steps=1, multi_step_stream_outputs=True, scheduler_delay_factor=0.0, enable_chunked_prefill=None, speculative_model=None, speculative_model_quantization=None, num_speculative_tokens=None, speculative_disable_mqa_scorer=False, speculative_draft_tensor_parallel_size=None, speculative_max_model_len=None, speculative_disable_by_batch_size=None, ngram_prompt_lookup_max=None, ngram_prompt_lookup_min=None, spec_decoding_acceptance_method='rejection_sampler', typical_acceptance_sampler_posterior_threshold=None, typical_acceptance_sampler_posterior_alpha=None, disable_logprobs_during_spec_decoding=None, model_loader_extra_config=None, ignore_patterns=[], preemption_mode=None, served_model_name=None, qlora_adapter_name_or_path=None, otlp_traces_endpoint=None, collect_detailed_traces=None, disable_async_output_proc=False, scheduling_policy='fcfs', scheduler_cls='vllm.core.scheduler.Scheduler', override_neuron_config=None, override_pooler_config=None, compilation_config=None, kv_transfer_config=None, worker_cls='auto', generation_config=None, override_generation_config=None, enable_sleep_mode=False, calculate_kv_scales=False, additional_config=None, disable_log_requests=False, max_log_len=None, disable_fastapi_docs=False, enable_prompt_tokens_details=False, dispatch_function=<function ServeSubcommand.cmd at 0x7f30970ee4d0>)
INFO 06-24 16:50:05 api_server.py:209] Started engine process with PID 371
INFO 06-24 16:50:08 __init__.py:30] Available plugins for group vllm.platform_plugins:
INFO 06-24 16:50:08 __init__.py:32] name=ascend, value=vllm_ascend:register
INFO 06-24 16:50:08 __init__.py:34] all available plugins for group vllm.platform_plugins will be loaded.
INFO 06-24 16:50:08 __init__.py:36] set environment variable VLLM_PLUGINS to control which plugins to load.
INFO 06-24 16:50:08 __init__.py:44] plugin ascend loaded.
INFO 06-24 16:50:08 __init__.py:198] Platform plugin ascend is activated
WARNING:root:Warning: Failed to register custom ops, all custom ops will be disabled
INFO 06-24 16:50:09 __init__.py:30] Available plugins for group vllm.general_plugins:
INFO 06-24 16:50:09 __init__.py:32] name=ascend_enhanced_model, value=vllm_ascend:register_model
INFO 06-24 16:50:09 __init__.py:34] all available plugins for group vllm.general_plugins will be loaded.
INFO 06-24 16:50:09 __init__.py:36] set environment variable VLLM_PLUGINS to control which plugins to load.
INFO 06-24 16:50:09 __init__.py:44] plugin ascend_enhanced_model loaded.
WARNING 06-24 16:50:09 registry.py:351] Model architecture Qwen2VLForConditionalGeneration is already registered, and will be overwritten by the new model class vllm_ascend.models.qwen2_vl:AscendQwen2VLForConditionalGeneration.
WARNING 06-24 16:50:09 registry.py:351] Model architecture Qwen2_5_VLForConditionalGeneration is already registered, and will be overwritten by the new model class vllm_ascend.models.qwen2_5_vl:AscendQwen2_5_VLForConditionalGeneration.
WARNING 06-24 16:50:09 registry.py:351] Model architecture DeepseekV2ForCausalLM is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_v2:CustomDeepseekV2ForCausalLM.
WARNING 06-24 16:50:09 registry.py:351] Model architecture DeepseekV3ForCausalLM is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_v2:CustomDeepseekV3ForCausalLM.
WARNING 06-24 16:50:09 registry.py:351] Model architecture DeepSeekMTPModel is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_mtp:CustomDeepSeekMTP.
INFO 06-24 16:50:09 importing.py:16] Triton not installed or not compatible; certain GPU-related functions will not be available.
INFO 06-24 16:50:10 config.py:549] This model supports multiple tasks: {'score', 'classify', 'embed', 'reward', 'generate'}. Defaulting to 'generate'.
WARNING 06-24 16:50:10 arg_utils.py:1197] The model has a long context length (131072). This may cause OOM errors during the initial memory profiling phase, or result in low performance due to small KV cache space. Consider setting --max-model-len to a smaller value.
INFO 06-24 16:50:13 config.py:549] This model supports multiple tasks: {'embed', 'score', 'reward', 'classify', 'generate'}. Defaulting to 'generate'.
WARNING 06-24 16:50:13 arg_utils.py:1197] The model has a long context length (131072). This may cause OOM errors during the initial memory profiling phase, or result in low performance due to small KV cache space. Consider setting --max-model-len to a smaller value.
INFO 06-24 16:50:14 llm_engine.py:234] Initializing a V0 LLM engine (v0.7.3) with config: model='/home/models/Qwen2.5-7B', speculative_config=None, tokenizer='/home/models/Qwen2.5-7B', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, override_neuron_config=None, tokenizer_revision=None, trust_remote_code=False, dtype=torch.bfloat16, max_seq_len=131072, download_dir=None, load_format=auto, tensor_parallel_size=1, pipeline_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto, device_config=npu, decoding_config=DecodingConfig(guided_decoding_backend='xgrammar'), observability_config=ObservabilityConfig(otlp_traces_endpoint=None, collect_model_forward_time=False, collect_model_execute_time=False), seed=0, served_model_name=/home/models/Qwen2.5-7B, num_scheduler_steps=1, multi_step_stream_outputs=True, enable_prefix_caching=False, chunked_prefill_enabled=False, use_async_output_proc=True, disable_mm_preprocessor_cache=False, mm_processor_kwargs=None, pooler_config=None, compilation_config={"splitting_ops":[],"compile_sizes":[],"cudagraph_capture_sizes":[256,248,240,232,224,216,208,200,192,184,176,168,160,152,144,136,128,120,112,104,96,88,80,72,64,56,48,40,32,24,16,8,4,2,1],"max_capture_size":256}, use_cached_outputs=True,
WARNING 06-24 16:50:14 camem.py:69] Failed to import vllm_ascend_C:No module named 'vllm_ascend.vllm_ascend_C'
/usr/local/python3.10.17/lib/python3.10/site-packages/torch_npu/contrib/transfer_to_npu.py:292: ImportWarning:
*************************************************************************************************************
The torch.Tensor.cuda and torch.nn.Module.cuda are replaced with torch.Tensor.npu and torch.nn.Module.npu now..
The torch.cuda.DoubleTensor is replaced with torch.npu.FloatTensor cause the double type is not supported now..
The backend in torch.distributed.init_process_group set to hccl now..
The torch.cuda.* and torch.cuda.amp.* are replaced with torch.npu.* and torch.npu.amp.* now..
The device parameters have been replaced with npu in the function below:
torch.logspace, torch.randint, torch.hann_window, torch.rand, torch.full_like, torch.ones_like, torch.rand_like, torch.randperm, torch.arange, torch.frombuffer, torch.normal, torch._empty_per_channel_affine_quantized, torch.empty_strided, torch.empty_like, torch.scalar_tensor, torch.tril_indices, torch.bartlett_window, torch.ones, torch.sparse_coo_tensor, torch.randn, torch.kaiser_window, torch.tensor, torch.triu_indices, torch.as_tensor, torch.zeros, torch.randint_like, torch.full, torch.eye, torch._sparse_csr_tensor_unsafe, torch.empty, torch._sparse_coo_tensor_unsafe, torch.blackman_window, torch.zeros_like, torch.range, torch.sparse_csr_tensor, torch.randn_like, torch.from_file, torch._cudnn_init_dropout_state, torch._empty_affine_quantized, torch.linspace, torch.hamming_window, torch.empty_quantized, torch._pin_memory, torch.autocast, torch.load, torch.Generator, torch.set_default_device, torch.Tensor.new_empty, torch.Tensor.new_empty_strided, torch.Tensor.new_full, torch.Tensor.new_ones, torch.Tensor.new_tensor, torch.Tensor.new_zeros, torch.Tensor.to, torch.Tensor.pin_memory, torch.nn.Module.to, torch.nn.Module.to_empty
*************************************************************************************************************
warnings.warn(msg, ImportWarning)
/usr/local/python3.10.17/lib/python3.10/site-packages/torch_npu/contrib/transfer_to_npu.py:247: RuntimeWarning: torch.jit.script and torch.jit.script_method will be disabled by transfer_to_npu, which currently does not support them, if you need to enable them, please do not use transfer_to_npu.
warnings.warn(msg, RuntimeWarning)
WARNING 06-24 16:50:14 utils.py:2262] Methods add_prompt_adapter,cache_config,compilation_config,current_platform,list_prompt_adapters,load_config,pin_prompt_adapter,remove_prompt_adapter not implemented in <vllm_ascend.worker.worker.NPUWorker object at 0x7fef3a077940>
INFO 06-24 16:50:20 model_runner.py:902] Starting to load model /home/models/Qwen2.5-7B...
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
ERROR 06-24 16:50:35 engine.py:400] Error while deserializing header: HeaderTooLarge
ERROR 06-24 16:50:35 engine.py:400] Traceback (most recent call last):
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 391, in run_mp_engine
ERROR 06-24 16:50:35 engine.py:400] engine = MQLLMEngine.from_engine_args(engine_args=engine_args,
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 124, in from_engine_args
ERROR 06-24 16:50:35 engine.py:400] return cls(ipc_path=ipc_path,
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 76, in __init__
ERROR 06-24 16:50:35 engine.py:400] self.engine = LLMEngine(*args, **kwargs)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/engine/llm_engine.py", line 273, in __init__
ERROR 06-24 16:50:35 engine.py:400] self.model_executor = executor_class(vllm_config=vllm_config, )
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/executor/executor_base.py", line 52, in __init__
ERROR 06-24 16:50:35 engine.py:400] self._init_executor()
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/executor/uniproc_executor.py", line 47, in _init_executor
ERROR 06-24 16:50:35 engine.py:400] self.collective_rpc("load_model")
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/executor/uniproc_executor.py", line 56, in collective_rpc
ERROR 06-24 16:50:35 engine.py:400] answer = run_method(self.driver_worker, method, args, kwargs)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/utils.py", line 2196, in run_method
ERROR 06-24 16:50:35 engine.py:400] return func(*args, **kwargs)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/worker.py", line 207, in load_model
ERROR 06-24 16:50:35 engine.py:400] self.model_runner.load_model()
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/model_runner.py", line 904, in load_model
ERROR 06-24 16:50:35 engine.py:400] self.model = get_model(vllm_config=self.vllm_config)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/__init__.py", line 14, in get_model
ERROR 06-24 16:50:35 engine.py:400] return loader.load_model(vllm_config=vllm_config)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 409, in load_model
ERROR 06-24 16:50:35 engine.py:400] loaded_weights = model.load_weights(
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/qwen2.py", line 515, in load_weights
ERROR 06-24 16:50:35 engine.py:400] return loader.load_weights(weights)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 235, in load_weights
ERROR 06-24 16:50:35 engine.py:400] autoloaded_weights = set(self._load_module("", self.module, weights))
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 187, in _load_module
ERROR 06-24 16:50:35 engine.py:400] for child_prefix, child_weights in self._groupby_prefix(weights):
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 101, in _groupby_prefix
ERROR 06-24 16:50:35 engine.py:400] for prefix, group in itertools.groupby(weights_by_parts,
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 98, in <genexpr>
ERROR 06-24 16:50:35 engine.py:400] weights_by_parts = ((weight_name.split(".", 1), weight_data)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 385, in _get_all_weights
ERROR 06-24 16:50:35 engine.py:400] yield from self._get_weights_iterator(primary_weights)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 368, in <genexpr>
ERROR 06-24 16:50:35 engine.py:400] return ((source.prefix + name, tensor)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/weight_utils.py", line 425, in safetensors_weights_iterator
ERROR 06-24 16:50:35 engine.py:400] with safe_open(st_file, framework="pt") as f:
ERROR 06-24 16:50:35 engine.py:400] safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge
Process SpawnProcess-1:
Traceback (most recent call last):
File "/usr/local/python3.10.17/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap
self.run()
File "/usr/local/python3.10.17/lib/python3.10/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 402, in run_mp_engine
raise e
File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 391, in run_mp_engine
engine = MQLLMEngine.from_engine_args(engine_args=engine_args,
File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 124, in from_engine_args
return cls(ipc_path=ipc_path,
File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 76, in __init__
self.engine = LLMEngine(*args, **kwargs)
File "/vllm-workspace/vllm/vllm/engine/llm_engine.py", line 273, in __init__
self.model_executor = executor_class(vllm_config=vllm_config, )
File "/vllm-workspace/vllm/vllm/executor/executor_base.py", line 52, in __init__
self._init_executor()
File "/vllm-workspace/vllm/vllm/executor/uniproc_executor.py", line 47, in _init_executor
self.collective_rpc("load_model")
File "/vllm-workspace/vllm/vllm/executor/uniproc_executor.py", line 56, in collective_rpc
answer = run_method(self.driver_worker, method, args, kwargs)
File "/vllm-workspace/vllm/vllm/utils.py", line 2196, in run_method
return func(*args, **kwargs)
File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/worker.py", line 207, in load_model
self.model_runner.load_model()
File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/model_runner.py", line 904, in load_model
self.model = get_model(vllm_config=self.vllm_config)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/__init__.py", line 14, in get_model
return loader.load_model(vllm_config=vllm_config)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 409, in load_model
loaded_weights = model.load_weights(
File "/vllm-workspace/vllm/vllm/model_executor/models/qwen2.py", line 515, in load_weights
return loader.load_weights(weights)
File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 235, in load_weights
autoloaded_weights = set(self._load_module("", self.module, weights))
File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 187, in _load_module
for child_prefix, child_weights in self._groupby_prefix(weights):
File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 101, in _groupby_prefix
for prefix, group in itertools.groupby(weights_by_parts,
File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 98, in <genexpr>
weights_by_parts = ((weight_name.split(".", 1), weight_data)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 385, in _get_all_weights
yield from self._get_weights_iterator(primary_weights)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 368, in <genexpr>
return ((source.prefix + name, tensor)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/weight_utils.py", line 425, in safetensors_weights_iterator
with safe_open(st_file, framework="pt") as f:
safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge
Traceback (most recent call last):
File "/usr/local/python3.10.17/bin/vllm", line 8, in <module>
sys.exit(main())
File "/vllm-workspace/vllm/vllm/entrypoints/cli/main.py", line 73, in main
args.dispatch_function(args)
File "/vllm-workspace/vllm/vllm/entrypoints/cli/serve.py", line 34, in cmd
uvloop.run(run_server(args))
File "/usr/local/python3.10.17/lib/python3.10/site-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/python3.10.17/lib/python3.10/site-packages/uvloop/__init__.py", line 61, in wrapper
return await main
File "/vllm-workspace/vllm/vllm/entrypoints/openai/api_server.py", line 947, in run_server
async with build_async_engine_client(args) as engine_client:
File "/usr/local/python3.10.17/lib/python3.10/contextlib.py", line 199, in __aenter__
return await anext(self.gen)
File "/vllm-workspace/vllm/vllm/entrypoints/openai/api_server.py", line 139, in build_async_engine_client
async with build_async_engine_client_from_engine_args(
File "/usr/local/python3.10.17/lib/python3.10/contextlib.py", line 199, in __aenter__
return await anext(self.gen)
File "/vllm-workspace/vllm/vllm/entrypoints/openai/api_server.py", line 233, in build_async_engine_client_from_engine_args
raise RuntimeError(
RuntimeError: Engine process failed to start. See stack trace for the root cause.
[ERROR] 2025-06-24-16:50:41 (PID:346, Device:-1, RankID:-1) ERR99999 UNKNOWN applicaiton exception
点赞
收藏
回复
分享
举报
已解决
本来想提交工单，但是提了八次都提示我提交失败，只能发帖求助o(╥﹏╥)o
我安装了docker版的vllm-ascend，启动大模型时报错，错误日志如下
驱动版本为25.0.rc1.1，固件版本为7.7.0.1.231, 均安装正常，系统为欧拉22.03，docker版本为24.0.8,安装的镜像为quay.io/ascend/vllm-ascend:v0.7.3-dev-openeuler
创建容器命令为
docker run -itd \
--name vllm-ascend-env \
--device /dev/davinci0 \
--device /dev/davinci1 \
--device /dev/davinci_manager \
--device /dev/devmm_svm \
--device /dev/hisi_hdc \
-v /usr/local/dcmi:/usr/local/dcmi \
-v /usr/local/bin/npu-smi:/usr/local/bin/npu-smi \
-v /usr/local/Ascend/driver/lib64/:/usr/local/Ascend/driver/lib64/ \
-v /usr/local/Ascend/driver/version.info:/usr/local/Ascend/driver/version.info \
-v /etc/ascend_install.info:/etc/ascend_install.info \
-v /home/models:/home/models \
-it quay.io/ascend/vllm-ascend:v0.7.3-dev-openeuler bash
启动大模型时报错 ，日志如下：
[root@09be6b1eee4f models]# vllm serve /home/models/Qwen2.5-7B --dtype auto --port 8000
INFO 06-24 16:50:05 __init__.py:30] Available plugins for group vllm.platform_plugins:
INFO 06-24 16:50:05 __init__.py:32] name=ascend, value=vllm_ascend:register
INFO 06-24 16:50:05 __init__.py:34] all available plugins for group vllm.platform_plugins will be loaded.
INFO 06-24 16:50:05 __init__.py:36] set environment variable VLLM_PLUGINS to control which plugins to load.
INFO 06-24 16:50:05 __init__.py:44] plugin ascend loaded.
INFO 06-24 16:50:05 __init__.py:198] Platform plugin ascend is activated
WARNING:root:Warning: Failed to register custom ops, all custom ops will be disabled
INFO 06-24 16:50:05 __init__.py:30] Available plugins for group vllm.general_plugins:
INFO 06-24 16:50:05 __init__.py:32] name=ascend_enhanced_model, value=vllm_ascend:register_model
INFO 06-24 16:50:05 __init__.py:34] all available plugins for group vllm.general_plugins will be loaded.
INFO 06-24 16:50:05 __init__.py:36] set environment variable VLLM_PLUGINS to control which plugins to load.
INFO 06-24 16:50:05 __init__.py:44] plugin ascend_enhanced_model loaded.
WARNING 06-24 16:50:05 registry.py:351] Model architecture Qwen2VLForConditionalGeneration is already registered, and will be overwritten by the new model class vllm_ascend.models.qwen2_vl:AscendQwen2VLForConditionalGeneration.
WARNING 06-24 16:50:05 registry.py:351] Model architecture Qwen2_5_VLForConditionalGeneration is already registered, and will be overwritten by the new model class vllm_ascend.models.qwen2_5_vl:AscendQwen2_5_VLForConditionalGeneration.
WARNING 06-24 16:50:05 registry.py:351] Model architecture DeepseekV2ForCausalLM is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_v2:CustomDeepseekV2ForCausalLM.
WARNING 06-24 16:50:05 registry.py:351] Model architecture DeepseekV3ForCausalLM is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_v2:CustomDeepseekV3ForCausalLM.
WARNING 06-24 16:50:05 registry.py:351] Model architecture DeepSeekMTPModel is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_mtp:CustomDeepSeekMTP.
INFO 06-24 16:50:05 importing.py:16] Triton not installed or not compatible; certain GPU-related functions will not be available.
INFO 06-24 16:50:05 api_server.py:912] vLLM API server version 0.7.3
INFO 06-24 16:50:05 api_server.py:913] args: Namespace(subparser='serve', model_tag='/home/models/Qwen2.5-7B', config='', host=None, port=8000, uvicorn_log_level='info', allow_credentials=False, allowed_origins=['*'], allowed_methods=['*'], allowed_headers=['*'], api_key=None, lora_modules=None, prompt_adapters=None, chat_template=None, chat_template_content_format='auto', response_role='assistant', ssl_keyfile=None, ssl_certfile=None, ssl_ca_certs=None, ssl_cert_reqs=0, root_path=None, middleware=[], return_tokens_as_token_ids=False, disable_frontend_multiprocessing=False, enable_request_id_headers=False, enable_auto_tool_choice=False, enable_reasoning=False, reasoning_parser=None, tool_call_parser=None, tool_parser_plugin='', model='/home/models/Qwen2.5-7B', task='auto', tokenizer=None, skip_tokenizer_init=False, revision=None, code_revision=None, tokenizer_revision=None, tokenizer_mode='auto', trust_remote_code=False, allowed_local_media_path=None, download_dir=None, load_format='auto', config_format=<ConfigFormat.AUTO: 'auto'>, dtype='auto', kv_cache_dtype='auto', max_model_len=None, guided_decoding_backend='xgrammar', logits_processor_pattern=None, model_impl='auto', distributed_executor_backend=None, pipeline_parallel_size=1, tensor_parallel_size=1, max_parallel_loading_workers=None, ray_workers_use_nsight=False, block_size=None, enable_prefix_caching=None, disable_sliding_window=False, use_v2_block_manager=True, num_lookahead_slots=0, seed=0, swap_space=4, cpu_offload_gb=0, gpu_memory_utilization=0.9, num_gpu_blocks_override=None, max_num_batched_tokens=None, max_num_partial_prefills=1, max_long_partial_prefills=1, long_prefill_token_threshold=0, max_num_seqs=None, max_logprobs=20, disable_log_stats=False, quantization=None, rope_scaling=None, rope_theta=None, hf_overrides=None, enforce_eager=False, max_seq_len_to_capture=8192, disable_custom_all_reduce=False, tokenizer_pool_size=0, tokenizer_pool_type='ray', tokenizer_pool_extra_config=None, limit_mm_per_prompt=None, mm_processor_kwargs=None, disable_mm_preprocessor_cache=False, enable_lora=False, enable_lora_bias=False, max_loras=1, max_lora_rank=16, lora_extra_vocab_size=256, lora_dtype='auto', long_lora_scaling_factors=None, max_cpu_loras=None, fully_sharded_loras=False, enable_prompt_adapter=False, max_prompt_adapters=1, max_prompt_adapter_token=0, device='auto', num_scheduler_steps=1, multi_step_stream_outputs=True, scheduler_delay_factor=0.0, enable_chunked_prefill=None, speculative_model=None, speculative_model_quantization=None, num_speculative_tokens=None, speculative_disable_mqa_scorer=False, speculative_draft_tensor_parallel_size=None, speculative_max_model_len=None, speculative_disable_by_batch_size=None, ngram_prompt_lookup_max=None, ngram_prompt_lookup_min=None, spec_decoding_acceptance_method='rejection_sampler', typical_acceptance_sampler_posterior_threshold=None, typical_acceptance_sampler_posterior_alpha=None, disable_logprobs_during_spec_decoding=None, model_loader_extra_config=None, ignore_patterns=[], preemption_mode=None, served_model_name=None, qlora_adapter_name_or_path=None, otlp_traces_endpoint=None, collect_detailed_traces=None, disable_async_output_proc=False, scheduling_policy='fcfs', scheduler_cls='vllm.core.scheduler.Scheduler', override_neuron_config=None, override_pooler_config=None, compilation_config=None, kv_transfer_config=None, worker_cls='auto', generation_config=None, override_generation_config=None, enable_sleep_mode=False, calculate_kv_scales=False, additional_config=None, disable_log_requests=False, max_log_len=None, disable_fastapi_docs=False, enable_prompt_tokens_details=False, dispatch_function=<function ServeSubcommand.cmd at 0x7f30970ee4d0>)
INFO 06-24 16:50:05 api_server.py:209] Started engine process with PID 371
INFO 06-24 16:50:08 __init__.py:30] Available plugins for group vllm.platform_plugins:
INFO 06-24 16:50:08 __init__.py:32] name=ascend, value=vllm_ascend:register
INFO 06-24 16:50:08 __init__.py:34] all available plugins for group vllm.platform_plugins will be loaded.
INFO 06-24 16:50:08 __init__.py:36] set environment variable VLLM_PLUGINS to control which plugins to load.
INFO 06-24 16:50:08 __init__.py:44] plugin ascend loaded.
INFO 06-24 16:50:08 __init__.py:198] Platform plugin ascend is activated
WARNING:root:Warning: Failed to register custom ops, all custom ops will be disabled
INFO 06-24 16:50:09 __init__.py:30] Available plugins for group vllm.general_plugins:
INFO 06-24 16:50:09 __init__.py:32] name=ascend_enhanced_model, value=vllm_ascend:register_model
INFO 06-24 16:50:09 __init__.py:34] all available plugins for group vllm.general_plugins will be loaded.
INFO 06-24 16:50:09 __init__.py:36] set environment variable VLLM_PLUGINS to control which plugins to load.
INFO 06-24 16:50:09 __init__.py:44] plugin ascend_enhanced_model loaded.
WARNING 06-24 16:50:09 registry.py:351] Model architecture Qwen2VLForConditionalGeneration is already registered, and will be overwritten by the new model class vllm_ascend.models.qwen2_vl:AscendQwen2VLForConditionalGeneration.
WARNING 06-24 16:50:09 registry.py:351] Model architecture Qwen2_5_VLForConditionalGeneration is already registered, and will be overwritten by the new model class vllm_ascend.models.qwen2_5_vl:AscendQwen2_5_VLForConditionalGeneration.
WARNING 06-24 16:50:09 registry.py:351] Model architecture DeepseekV2ForCausalLM is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_v2:CustomDeepseekV2ForCausalLM.
WARNING 06-24 16:50:09 registry.py:351] Model architecture DeepseekV3ForCausalLM is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_v2:CustomDeepseekV3ForCausalLM.
WARNING 06-24 16:50:09 registry.py:351] Model architecture DeepSeekMTPModel is already registered, and will be overwritten by the new model class vllm_ascend.models.deepseek_mtp:CustomDeepSeekMTP.
INFO 06-24 16:50:09 importing.py:16] Triton not installed or not compatible; certain GPU-related functions will not be available.
INFO 06-24 16:50:10 config.py:549] This model supports multiple tasks: {'score', 'classify', 'embed', 'reward', 'generate'}. Defaulting to 'generate'.
WARNING 06-24 16:50:10 arg_utils.py:1197] The model has a long context length (131072). This may cause OOM errors during the initial memory profiling phase, or result in low performance due to small KV cache space. Consider setting --max-model-len to a smaller value.
INFO 06-24 16:50:13 config.py:549] This model supports multiple tasks: {'embed', 'score', 'reward', 'classify', 'generate'}. Defaulting to 'generate'.
WARNING 06-24 16:50:13 arg_utils.py:1197] The model has a long context length (131072). This may cause OOM errors during the initial memory profiling phase, or result in low performance due to small KV cache space. Consider setting --max-model-len to a smaller value.
INFO 06-24 16:50:14 llm_engine.py:234] Initializing a V0 LLM engine (v0.7.3) with config: model='/home/models/Qwen2.5-7B', speculative_config=None, tokenizer='/home/models/Qwen2.5-7B', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, override_neuron_config=None, tokenizer_revision=None, trust_remote_code=False, dtype=torch.bfloat16, max_seq_len=131072, download_dir=None, load_format=auto, tensor_parallel_size=1, pipeline_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto, device_config=npu, decoding_config=DecodingConfig(guided_decoding_backend='xgrammar'), observability_config=ObservabilityConfig(otlp_traces_endpoint=None, collect_model_forward_time=False, collect_model_execute_time=False), seed=0, served_model_name=/home/models/Qwen2.5-7B, num_scheduler_steps=1, multi_step_stream_outputs=True, enable_prefix_caching=False, chunked_prefill_enabled=False, use_async_output_proc=True, disable_mm_preprocessor_cache=False, mm_processor_kwargs=None, pooler_config=None, compilation_config={"splitting_ops":[],"compile_sizes":[],"cudagraph_capture_sizes":[256,248,240,232,224,216,208,200,192,184,176,168,160,152,144,136,128,120,112,104,96,88,80,72,64,56,48,40,32,24,16,8,4,2,1],"max_capture_size":256}, use_cached_outputs=True,
WARNING 06-24 16:50:14 camem.py:69] Failed to import vllm_ascend_C:No module named 'vllm_ascend.vllm_ascend_C'
/usr/local/python3.10.17/lib/python3.10/site-packages/torch_npu/contrib/transfer_to_npu.py:292: ImportWarning:
*************************************************************************************************************
The torch.Tensor.cuda and torch.nn.Module.cuda are replaced with torch.Tensor.npu and torch.nn.Module.npu now..
The torch.cuda.DoubleTensor is replaced with torch.npu.FloatTensor cause the double type is not supported now..
The backend in torch.distributed.init_process_group set to hccl now..
The torch.cuda.* and torch.cuda.amp.* are replaced with torch.npu.* and torch.npu.amp.* now..
The device parameters have been replaced with npu in the function below:
torch.logspace, torch.randint, torch.hann_window, torch.rand, torch.full_like, torch.ones_like, torch.rand_like, torch.randperm, torch.arange, torch.frombuffer, torch.normal, torch._empty_per_channel_affine_quantized, torch.empty_strided, torch.empty_like, torch.scalar_tensor, torch.tril_indices, torch.bartlett_window, torch.ones, torch.sparse_coo_tensor, torch.randn, torch.kaiser_window, torch.tensor, torch.triu_indices, torch.as_tensor, torch.zeros, torch.randint_like, torch.full, torch.eye, torch._sparse_csr_tensor_unsafe, torch.empty, torch._sparse_coo_tensor_unsafe, torch.blackman_window, torch.zeros_like, torch.range, torch.sparse_csr_tensor, torch.randn_like, torch.from_file, torch._cudnn_init_dropout_state, torch._empty_affine_quantized, torch.linspace, torch.hamming_window, torch.empty_quantized, torch._pin_memory, torch.autocast, torch.load, torch.Generator, torch.set_default_device, torch.Tensor.new_empty, torch.Tensor.new_empty_strided, torch.Tensor.new_full, torch.Tensor.new_ones, torch.Tensor.new_tensor, torch.Tensor.new_zeros, torch.Tensor.to, torch.Tensor.pin_memory, torch.nn.Module.to, torch.nn.Module.to_empty
*************************************************************************************************************
warnings.warn(msg, ImportWarning)
/usr/local/python3.10.17/lib/python3.10/site-packages/torch_npu/contrib/transfer_to_npu.py:247: RuntimeWarning: torch.jit.script and torch.jit.script_method will be disabled by transfer_to_npu, which currently does not support them, if you need to enable them, please do not use transfer_to_npu.
warnings.warn(msg, RuntimeWarning)
WARNING 06-24 16:50:14 utils.py:2262] Methods add_prompt_adapter,cache_config,compilation_config,current_platform,list_prompt_adapters,load_config,pin_prompt_adapter,remove_prompt_adapter not implemented in <vllm_ascend.worker.worker.NPUWorker object at 0x7fef3a077940>
INFO 06-24 16:50:20 model_runner.py:902] Starting to load model /home/models/Qwen2.5-7B...
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
ERROR 06-24 16:50:35 engine.py:400] Error while deserializing header: HeaderTooLarge
ERROR 06-24 16:50:35 engine.py:400] Traceback (most recent call last):
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 391, in run_mp_engine
ERROR 06-24 16:50:35 engine.py:400] engine = MQLLMEngine.from_engine_args(engine_args=engine_args,
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 124, in from_engine_args
ERROR 06-24 16:50:35 engine.py:400] return cls(ipc_path=ipc_path,
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 76, in __init__
ERROR 06-24 16:50:35 engine.py:400] self.engine = LLMEngine(*args, **kwargs)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/engine/llm_engine.py", line 273, in __init__
ERROR 06-24 16:50:35 engine.py:400] self.model_executor = executor_class(vllm_config=vllm_config, )
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/executor/executor_base.py", line 52, in __init__
ERROR 06-24 16:50:35 engine.py:400] self._init_executor()
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/executor/uniproc_executor.py", line 47, in _init_executor
ERROR 06-24 16:50:35 engine.py:400] self.collective_rpc("load_model")
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/executor/uniproc_executor.py", line 56, in collective_rpc
ERROR 06-24 16:50:35 engine.py:400] answer = run_method(self.driver_worker, method, args, kwargs)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/utils.py", line 2196, in run_method
ERROR 06-24 16:50:35 engine.py:400] return func(*args, **kwargs)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/worker.py", line 207, in load_model
ERROR 06-24 16:50:35 engine.py:400] self.model_runner.load_model()
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/model_runner.py", line 904, in load_model
ERROR 06-24 16:50:35 engine.py:400] self.model = get_model(vllm_config=self.vllm_config)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/__init__.py", line 14, in get_model
ERROR 06-24 16:50:35 engine.py:400] return loader.load_model(vllm_config=vllm_config)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 409, in load_model
ERROR 06-24 16:50:35 engine.py:400] loaded_weights = model.load_weights(
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/qwen2.py", line 515, in load_weights
ERROR 06-24 16:50:35 engine.py:400] return loader.load_weights(weights)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 235, in load_weights
ERROR 06-24 16:50:35 engine.py:400] autoloaded_weights = set(self._load_module("", self.module, weights))
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 187, in _load_module
ERROR 06-24 16:50:35 engine.py:400] for child_prefix, child_weights in self._groupby_prefix(weights):
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 101, in _groupby_prefix
ERROR 06-24 16:50:35 engine.py:400] for prefix, group in itertools.groupby(weights_by_parts,
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 98, in <genexpr>
ERROR 06-24 16:50:35 engine.py:400] weights_by_parts = ((weight_name.split(".", 1), weight_data)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 385, in _get_all_weights
ERROR 06-24 16:50:35 engine.py:400] yield from self._get_weights_iterator(primary_weights)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 368, in <genexpr>
ERROR 06-24 16:50:35 engine.py:400] return ((source.prefix + name, tensor)
ERROR 06-24 16:50:35 engine.py:400] File "/vllm-workspace/vllm/vllm/model_executor/model_loader/weight_utils.py", line 425, in safetensors_weights_iterator
ERROR 06-24 16:50:35 engine.py:400] with safe_open(st_file, framework="pt") as f:
ERROR 06-24 16:50:35 engine.py:400] safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge
Process SpawnProcess-1:
Traceback (most recent call last):
File "/usr/local/python3.10.17/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap
self.run()
File "/usr/local/python3.10.17/lib/python3.10/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 402, in run_mp_engine
raise e
File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 391, in run_mp_engine
engine = MQLLMEngine.from_engine_args(engine_args=engine_args,
File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 124, in from_engine_args
return cls(ipc_path=ipc_path,
File "/vllm-workspace/vllm/vllm/engine/multiprocessing/engine.py", line 76, in __init__
self.engine = LLMEngine(*args, **kwargs)
File "/vllm-workspace/vllm/vllm/engine/llm_engine.py", line 273, in __init__
self.model_executor = executor_class(vllm_config=vllm_config, )
File "/vllm-workspace/vllm/vllm/executor/executor_base.py", line 52, in __init__
self._init_executor()
File "/vllm-workspace/vllm/vllm/executor/uniproc_executor.py", line 47, in _init_executor
self.collective_rpc("load_model")
File "/vllm-workspace/vllm/vllm/executor/uniproc_executor.py", line 56, in collective_rpc
answer = run_method(self.driver_worker, method, args, kwargs)
File "/vllm-workspace/vllm/vllm/utils.py", line 2196, in run_method
return func(*args, **kwargs)
File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/worker.py", line 207, in load_model
self.model_runner.load_model()
File "/vllm-workspace/vllm-ascend/vllm_ascend/worker/model_runner.py", line 904, in load_model
self.model = get_model(vllm_config=self.vllm_config)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/__init__.py", line 14, in get_model
return loader.load_model(vllm_config=vllm_config)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 409, in load_model
loaded_weights = model.load_weights(
File "/vllm-workspace/vllm/vllm/model_executor/models/qwen2.py", line 515, in load_weights
return loader.load_weights(weights)
File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 235, in load_weights
autoloaded_weights = set(self._load_module("", self.module, weights))
File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 187, in _load_module
for child_prefix, child_weights in self._groupby_prefix(weights):
File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 101, in _groupby_prefix
for prefix, group in itertools.groupby(weights_by_parts,
File "/vllm-workspace/vllm/vllm/model_executor/models/utils.py", line 98, in <genexpr>
weights_by_parts = ((weight_name.split(".", 1), weight_data)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 385, in _get_all_weights
yield from self._get_weights_iterator(primary_weights)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/loader.py", line 368, in <genexpr>
return ((source.prefix + name, tensor)
File "/vllm-workspace/vllm/vllm/model_executor/model_loader/weight_utils.py", line 425, in safetensors_weights_iterator
with safe_open(st_file, framework="pt") as f:
safetensors_rust.SafetensorError: Error while deserializing header: HeaderTooLarge
Traceback (most recent call last):
File "/usr/local/python3.10.17/bin/vllm", line 8, in <module>
sys.exit(main())
File "/vllm-workspace/vllm/vllm/entrypoints/cli/main.py", line 73, in main
args.dispatch_function(args)
File "/vllm-workspace/vllm/vllm/entrypoints/cli/serve.py", line 34, in cmd
uvloop.run(run_server(args))
File "/usr/local/python3.10.17/lib/python3.10/site-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/python3.10.17/lib/python3.10/site-packages/uvloop/__init__.py", line 61, in wrapper
return await main
File "/vllm-workspace/vllm/vllm/entrypoints/openai/api_server.py", line 947, in run_server
async with build_async_engine_client(args) as engine_client:
File "/usr/local/python3.10.17/lib/python3.10/contextlib.py", line 199, in __aenter__
return await anext(self.gen)
File "/vllm-workspace/vllm/vllm/entrypoints/openai/api_server.py", line 139, in build_async_engine_client
async with build_async_engine_client_from_engine_args(
File "/usr/local/python3.10.17/lib/python3.10/contextlib.py", line 199, in __aenter__
return await anext(self.gen)
File "/vllm-workspace/vllm/vllm/entrypoints/openai/api_server.py", line 233, in build_async_engine_client_from_engine_args
raise RuntimeError(
RuntimeError: Engine process failed to start. See stack trace for the root cause.
[ERROR] 2025-06-24-16:50:41 (PID:346, Device:-1, RankID:-1) ERR99999 UNKNOWN applicaiton exception