vllm: [Bug]: Llama 4 EOFError

Your current environment

The output of `python collect_env.py`
root@b33811284aa6:/home/ndurkee# python3 collect_env.py
INFO 04-06 06:49:09 [__init__.py:239] Automatically detected platform cuda.
Collecting environment information...
PyTorch version: 2.6.0+cu124
Is debug build: False
CUDA used to build PyTorch: 12.4
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.4 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 4.0.0
Libc version: glibc-2.35

Python version: 3.12.9 (main, Feb  5 2025, 08:49:00) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-4.18.0-425.19.2.el8_7.x86_64-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.4.131
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration:
GPU 0: NVIDIA A100-SXM4-40GB
GPU 1: NVIDIA A100-SXM4-40GB
GPU 2: NVIDIA A100-SXM4-40GB
GPU 3: NVIDIA A100-SXM4-40GB

Nvidia driver version: 525.105.17
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Address sizes:                   43 bits physical, 48 bits virtual
Byte Order:                      Little Endian
CPU(s):                          256
On-line CPU(s) list:             0-255
Vendor ID:                       AuthenticAMD
Model name:                      AMD EPYC 7742 64-Core Processor
CPU family:                      23
Model:                           49
Thread(s) per core:              2
Core(s) per socket:              64
Socket(s):                       2
Stepping:                        0
Frequency boost:                 enabled
CPU max MHz:                     2250.0000
CPU min MHz:                     1500.0000
BogoMIPS:                        4491.45
Flags:                           fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr wbnoinvd arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip rdpid overflow_recov succor smca sme sev sev_es
Virtualization:                  AMD-V
L1d cache:                       4 MiB (128 instances)
L1i cache:                       4 MiB (128 instances)
L2 cache:                        64 MiB (128 instances)
L3 cache:                        512 MiB (32 instances)
NUMA node(s):                    8
NUMA node0 CPU(s):               0-15,128-143
NUMA node1 CPU(s):               16-31,144-159
NUMA node2 CPU(s):               32-47,160-175
NUMA node3 CPU(s):               48-63,176-191
NUMA node4 CPU(s):               64-79,192-207
NUMA node5 CPU(s):               80-95,208-223
NUMA node6 CPU(s):               96-111,224-239
NUMA node7 CPU(s):               112-127,240-255
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Mmio stale data:   Not affected
Vulnerability Retbleed:          Mitigation; untrained return thunk; SMT enabled with STIBP protection
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Mitigation; Retpolines, IBPB conditional, STIBP always-on, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected

Versions of relevant libraries:
[pip3] flashinfer-python==0.2.1.post2+cu124torch2.6
[pip3] numpy==2.1.3
[pip3] nvidia-cublas-cu12==12.4.5.8
[pip3] nvidia-cuda-cupti-cu12==12.4.127
[pip3] nvidia-cuda-nvrtc-cu12==12.4.127
[pip3] nvidia-cuda-runtime-cu12==12.4.127
[pip3] nvidia-cudnn-cu12==9.1.0.70
[pip3] nvidia-cufft-cu12==11.2.1.3
[pip3] nvidia-curand-cu12==10.3.5.147
[pip3] nvidia-cusolver-cu12==11.6.1.9
[pip3] nvidia-cusparse-cu12==12.3.1.170
[pip3] nvidia-cusparselt-cu12==0.6.2
[pip3] nvidia-nccl-cu12==2.21.5
[pip3] nvidia-nvjitlink-cu12==12.4.127
[pip3] nvidia-nvtx-cu12==12.4.127
[pip3] pyzmq==26.4.0
[pip3] torch==2.6.0
[pip3] torchaudio==2.6.0
[pip3] torchvision==0.21.0
[pip3] transformers==4.51.0
[pip3] triton==3.2.0
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.8.3
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    GPU1    GPU2    GPU3    NIC0    NIC1    NIC2    NIC3    NIC4    NIC5    NIC6    NIC7    NIC8    NIC9    CPU Affinity    NUMA Affinity
GPU0     X      NV12    NV12    NV12    PXB     PXB     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     48-63,176-191   3
GPU1    NV12     X      NV12    NV12    SYS     SYS     PXB     PXB     SYS     SYS     SYS     SYS     SYS     SYS     16-31,144-159   1
GPU2    NV12    NV12     X      NV12    SYS     SYS     SYS     SYS     PXB     PXB     SYS     SYS     SYS     SYS     112-127,240-255 7
GPU3    NV12    NV12    NV12     X      SYS     SYS     SYS     SYS     PXB     PXB     SYS     SYS     SYS     SYS     112-127,240-255 7
NIC0    PXB     SYS     SYS     SYS      X      PXB     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS
NIC1    PXB     SYS     SYS     SYS     PXB      X      SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS
NIC2    SYS     PXB     SYS     SYS     SYS     SYS      X      PXB     SYS     SYS     SYS     SYS     SYS     SYS
NIC3    SYS     PXB     SYS     SYS     SYS     SYS     PXB      X      SYS     SYS     SYS     SYS     SYS     SYS
NIC4    SYS     SYS     PXB     PXB     SYS     SYS     SYS     SYS      X      PXB     SYS     SYS     SYS     SYS
NIC5    SYS     SYS     PXB     PXB     SYS     SYS     SYS     SYS     PXB      X      SYS     SYS     SYS     SYS
NIC6    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      X      PXB     SYS     SYS
NIC7    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     PXB      X      SYS     SYS
NIC8    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      X      PIX
NIC9    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS     PIX      X

Legend:

  X    = Self
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PIX  = Connection traversing at most a single PCIe bridge
  NV#  = Connection traversing a bonded set of # NVLinks

NIC Legend:

  NIC0: mlx5_0
  NIC1: mlx5_1
  NIC2: mlx5_2
  NIC3: mlx5_3
  NIC4: mlx5_4
  NIC5: mlx5_5
  NIC6: mlx5_6
  NIC7: mlx5_7
  NIC8: mlx5_8
  NIC9: mlx5_9

NVIDIA_VISIBLE_DEVICES=1,3,4,5
NVIDIA_REQUIRE_CUDA=cuda>=12.4 brand=tesla,driver>=470,driver<471 brand=unknown,driver>=470,driver<471 brand=nvidia,driver>=470,driver<471 brand=nvidiartx,driver>=470,driver<471 brand=geforce,driver>=470,driver<471 brand=geforcertx,driver>=470,driver<471 brand=quadro,driver>=470,driver<471 brand=quadrortx,driver>=470,driver<471 brand=titan,driver>=470,driver<471 brand=titanrtx,driver>=470,driver<471 brand=tesla,driver>=525,driver<526 brand=unknown,driver>=525,driver<526 brand=nvidia,driver>=525,driver<526 brand=nvidiartx,driver>=525,driver<526 brand=geforce,driver>=525,driver<526 brand=geforcertx,driver>=525,driver<526 brand=quadro,driver>=525,driver<526 brand=quadrortx,driver>=525,driver<526 brand=titan,driver>=525,driver<526 brand=titanrtx,driver>=525,driver<526 brand=tesla,driver>=535,driver<536 brand=unknown,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=geforce,driver>=535,driver<536 brand=geforcertx,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=titan,driver>=535,driver<536 brand=titanrtx,driver>=535,driver<536
NCCL_VERSION=2.20.5-1
NVIDIA_DRIVER_CAPABILITIES=compute,utility
NVIDIA_PRODUCT_NAME=CUDA
VLLM_USAGE_SOURCE=production-docker-image
CUDA_VERSION=12.4.0
LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64
NCCL_CUMEM_ENABLE=0
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY

🐛 Describe the bug

Unable to run Llama 4 on 4xA100. Using 0.8.3 official docker image and the scout-instruct version from Meta

Using this command

sudo docker run --rm --name ndurkee_main --shm-size=10.24gb --gpus '"device=1,3,4,5"' -p 15010:15001 -p 15001:15001 -v /raid/ndurkee:/home/ndurkee arti.bsf.ball.com/docker-group/vllm/vllm-openai:v0.8.3 --model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ --max-model-len 32000 -tp 4 --gpu-memory-utilization 0.90 --port 15001 --max-log-len 10 --enable-prefix-caching

Note that I tried running this command with and without fp8 quantization and both failed. I also checked the safetensors files by themselves and they all loaded properly.

[ndurkee@AERO.BALL.COM@lscoaec-dgx0001 Llama-4-Scout-17B-16E-Instruct]$ sudo docker run --rm --name ndurkee_main --shm-size=10.24gb --gpus '"device=1,3,4,5"' -p 15010:15001 -p 15001:15001 -v /raid/ndurkee:/home/ndurkee arti.bsf.ball.com/docker-group/vllm/vllm-openai:v0.8.3 --model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ --max-model-len 32000 -tp 4 --gpu-memory-utilization 0.90 --port 15001 --max-log-len 10 --enable-prefix-caching
[sudo] password for ndurkee@AERO.BALL.COM:
INFO 04-06 06:44:35 [__init__.py:239] Automatically detected platform cuda.
INFO 04-06 06:44:38 [api_server.py:1034] vLLM API server version 0.8.3
INFO 04-06 06:44:38 [api_server.py:1035] args: Namespace(host=None, port=15001, uvicorn_log_level='info', disable_uvicorn_access_log=False, 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, enable_ssl_refresh=False, 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, tool_call_parser=None, tool_parser_plugin='', model='/home/ndurkee/Llama-4-Scout-17B-16E-Instruct/', task='auto', tokenizer=None, hf_config_path=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=32000, guided_decoding_backend='xgrammar', logits_processor_pattern=None, model_impl='auto', distributed_executor_backend=None, pipeline_parallel_size=1, tensor_parallel_size=4, data_parallel_size=1, enable_expert_parallel=False, max_parallel_loading_workers=None, ray_workers_use_nsight=False, block_size=None, enable_prefix_caching=True, prefix_caching_hash_algo='builtin', disable_sliding_window=False, use_v2_block_manager=True, num_lookahead_slots=0, seed=None, 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, use_tqdm_on_load=True, multi_step_stream_outputs=True, scheduler_delay_factor=0.0, enable_chunked_prefill=None, speculative_config=None, model_loader_extra_config=None, ignore_patterns=[], preemption_mode=None, served_model_name=None, qlora_adapter_name_or_path=None, show_hidden_metrics_for_version=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', worker_extension_cls='', generation_config='auto', override_generation_config=None, enable_sleep_mode=False, calculate_kv_scales=False, additional_config=None, enable_reasoning=False, reasoning_parser=None, disable_cascade_attn=False, disable_log_requests=False, max_log_len=10, disable_fastapi_docs=False, enable_prompt_tokens_details=False, enable_server_load_tracking=False)
INFO 04-06 06:44:44 [config.py:600] This model supports multiple tasks: {'generate', 'embed', 'reward', 'classify', 'score'}. Defaulting to 'generate'.
INFO 04-06 06:44:44 [config.py:1600] Defaulting to use mp for distributed inference
INFO 04-06 06:44:44 [config.py:1780] Chunked prefill is enabled with max_num_batched_tokens=2048.
INFO 04-06 06:44:49 [__init__.py:239] Automatically detected platform cuda.
INFO 04-06 06:44:51 [core.py:61] Initializing a V1 LLM engine (v0.8.3) with config: model='/home/ndurkee/Llama-4-Scout-17B-16E-Instruct/', speculative_config=None, tokenizer='/home/ndurkee/Llama-4-Scout-17B-16E-Instruct/', 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=32000, download_dir=None, load_format=LoadFormat.AUTO, tensor_parallel_size=4, pipeline_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto,  device_config=cuda, decoding_config=DecodingConfig(guided_decoding_backend='xgrammar', reasoning_backend=None), observability_config=ObservabilityConfig(show_hidden_metrics=False, otlp_traces_endpoint=None, collect_model_forward_time=False, collect_model_execute_time=False), seed=None, served_model_name=/home/ndurkee/Llama-4-Scout-17B-16E-Instruct/, num_scheduler_steps=1, multi_step_stream_outputs=True, enable_prefix_caching=True, chunked_prefill_enabled=True, use_async_output_proc=True, disable_mm_preprocessor_cache=False, mm_processor_kwargs=None, pooler_config=None, compilation_config={"level":3,"custom_ops":["none"],"splitting_ops":["vllm.unified_attention","vllm.unified_attention_with_output"],"use_inductor":true,"compile_sizes":[],"use_cudagraph":true,"cudagraph_num_of_warmups":1,"cudagraph_capture_sizes":[512,504,496,488,480,472,464,456,448,440,432,424,416,408,400,392,384,376,368,360,352,344,336,328,320,312,304,296,288,280,272,264,256,248,240,232,224,216,208,200,192,184,176,168,160,152,144,136,128,120,112,104,96,88,80,72,64,56,48,40,32,24,16,8,4,2,1],"max_capture_size":512}
WARNING 04-06 06:44:51 [multiproc_worker_utils.py:306] Reducing Torch parallelism from 128 threads to 1 to avoid unnecessary CPU contention. Set OMP_NUM_THREADS in the external environment to tune this value as needed.
INFO 04-06 06:44:51 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[0, 1, 2, 3], buffer_handle=(4, 10485760, 10, 'psm_451ec02c'), local_subscribe_addr='ipc:///tmp/0349da12-7899-4119-acf5-3c7bca857b78', remote_subscribe_addr=None, remote_addr_ipv6=False)
INFO 04-06 06:44:54 [__init__.py:239] Automatically detected platform cuda.
WARNING 04-06 06:44:58 [utils.py:2413] Methods determine_num_available_blocks,device_config,get_cache_block_size_bytes,initialize_cache not implemented in <vllm.v1.worker.gpu_worker.Worker object at 0x7f5d00dbe480>
(VllmWorker rank=0 pid=221) INFO 04-06 06:44:58 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[0], buffer_handle=(1, 10485760, 10, 'psm_e526fa01'), local_subscribe_addr='ipc:///tmp/4097c250-76b6-4073-83aa-a703633d034c', remote_subscribe_addr=None, remote_addr_ipv6=False)
INFO 04-06 06:45:01 [__init__.py:239] Automatically detected platform cuda.
WARNING 04-06 06:45:03 [utils.py:2413] Methods determine_num_available_blocks,device_config,get_cache_block_size_bytes,initialize_cache not implemented in <vllm.v1.worker.gpu_worker.Worker object at 0x7ff2c4096d20>
(VllmWorker rank=1 pid=238) INFO 04-06 06:45:03 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[0], buffer_handle=(1, 10485760, 10, 'psm_1f7d2e87'), local_subscribe_addr='ipc:///tmp/7d449291-5f2a-415e-9cdb-f805f1454d12', remote_subscribe_addr=None, remote_addr_ipv6=False)
INFO 04-06 06:45:07 [__init__.py:239] Automatically detected platform cuda.
WARNING 04-06 06:45:09 [utils.py:2413] Methods determine_num_available_blocks,device_config,get_cache_block_size_bytes,initialize_cache not implemented in <vllm.v1.worker.gpu_worker.Worker object at 0x7f952c09ab10>
(VllmWorker rank=2 pid=257) INFO 04-06 06:45:09 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[0], buffer_handle=(1, 10485760, 10, 'psm_50acc578'), local_subscribe_addr='ipc:///tmp/7dc49ecc-1706-4eba-94d0-389fbf0dc9c2', remote_subscribe_addr=None, remote_addr_ipv6=False)
INFO 04-06 06:45:12 [__init__.py:239] Automatically detected platform cuda.
WARNING 04-06 06:45:15 [utils.py:2413] Methods determine_num_available_blocks,device_config,get_cache_block_size_bytes,initialize_cache not implemented in <vllm.v1.worker.gpu_worker.Worker object at 0x7fab21f9f170>
(VllmWorker rank=3 pid=281) INFO 04-06 06:45:15 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[0], buffer_handle=(1, 10485760, 10, 'psm_bd618765'), local_subscribe_addr='ipc:///tmp/33c1e0cb-1870-449d-b57c-be834d857c47', remote_subscribe_addr=None, remote_addr_ipv6=False)
(VllmWorker rank=1 pid=238) INFO 04-06 06:45:15 [utils.py:990] Found nccl from library libnccl.so.2
(VllmWorker rank=3 pid=281) INFO 04-06 06:45:15 [utils.py:990] Found nccl from library libnccl.so.2
(VllmWorker rank=1 pid=238) INFO 04-06 06:45:15 [pynccl.py:69] vLLM is using nccl==2.21.5
(VllmWorker rank=3 pid=281) INFO 04-06 06:45:15 [pynccl.py:69] vLLM is using nccl==2.21.5
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:15 [utils.py:990] Found nccl from library libnccl.so.2
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:15 [pynccl.py:69] vLLM is using nccl==2.21.5
(VllmWorker rank=2 pid=257) INFO 04-06 06:45:15 [utils.py:990] Found nccl from library libnccl.so.2
(VllmWorker rank=2 pid=257) INFO 04-06 06:45:15 [pynccl.py:69] vLLM is using nccl==2.21.5
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:16 [custom_all_reduce_utils.py:206] generating GPU P2P access cache in /root/.cache/vllm/gpu_p2p_access_cache_for_0,1,2,3.json
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:40 [custom_all_reduce_utils.py:244] reading GPU P2P access cache from /root/.cache/vllm/gpu_p2p_access_cache_for_0,1,2,3.json
(VllmWorker rank=2 pid=257) INFO 04-06 06:45:40 [custom_all_reduce_utils.py:244] reading GPU P2P access cache from /root/.cache/vllm/gpu_p2p_access_cache_for_0,1,2,3.json
(VllmWorker rank=1 pid=238) INFO 04-06 06:45:40 [custom_all_reduce_utils.py:244] reading GPU P2P access cache from /root/.cache/vllm/gpu_p2p_access_cache_for_0,1,2,3.json
(VllmWorker rank=3 pid=281) INFO 04-06 06:45:40 [custom_all_reduce_utils.py:244] reading GPU P2P access cache from /root/.cache/vllm/gpu_p2p_access_cache_for_0,1,2,3.json
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:40 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[1, 2, 3], buffer_handle=(3, 4194304, 6, 'psm_7fad12d8'), local_subscribe_addr='ipc:///tmp/3d4c53c6-20c9-4481-a9e1-bde6fd832508', remote_subscribe_addr=None, remote_addr_ipv6=False)
(VllmWorker rank=2 pid=257) INFO 04-06 06:45:40 [parallel_state.py:957] rank 2 in world size 4 is assigned as DP rank 0, PP rank 0, TP rank 2
(VllmWorker rank=3 pid=281) INFO 04-06 06:45:40 [parallel_state.py:957] rank 3 in world size 4 is assigned as DP rank 0, PP rank 0, TP rank 3
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:40 [parallel_state.py:957] rank 0 in world size 4 is assigned as DP rank 0, PP rank 0, TP rank 0
(VllmWorker rank=1 pid=238) INFO 04-06 06:45:40 [parallel_state.py:957] rank 1 in world size 4 is assigned as DP rank 0, PP rank 0, TP rank 1
(VllmWorker rank=2 pid=257) INFO 04-06 06:45:40 [cuda.py:221] Using Flash Attention backend on V1 engine.
(VllmWorker rank=3 pid=281) INFO 04-06 06:45:40 [cuda.py:221] Using Flash Attention backend on V1 engine.
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:40 [cuda.py:221] Using Flash Attention backend on V1 engine.
(VllmWorker rank=1 pid=238) INFO 04-06 06:45:40 [cuda.py:221] Using Flash Attention backend on V1 engine.
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:44 [gpu_model_runner.py:1258] Starting to load model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/...
(VllmWorker rank=2 pid=257) INFO 04-06 06:45:44 [gpu_model_runner.py:1258] Starting to load model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/...
(VllmWorker rank=3 pid=281) INFO 04-06 06:45:44 [gpu_model_runner.py:1258] Starting to load model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/...
(VllmWorker rank=0 pid=221) INFO 04-06 06:45:45 [config.py:3334] cudagraph sizes specified by model runner [1, 2, 4, 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256, 264, 272, 280, 288, 296, 304, 312, 320, 328, 336, 344, 352, 360, 368, 376, 384, 392, 400, 408, 416, 424, 432, 440, 448, 456, 464, 472, 480, 488, 496, 504, 512] is overridden by config [512, 384, 256, 128, 4, 2, 1, 392, 264, 136, 8, 400, 272, 144, 16, 408, 280, 152, 24, 416, 288, 160, 32, 424, 296, 168, 40, 432, 304, 176, 48, 440, 312, 184, 56, 448, 320, 192, 64, 456, 328, 200, 72, 464, 336, 208, 80, 472, 344, 216, 88, 120, 480, 352, 248, 224, 96, 488, 504, 360, 232, 104, 496, 368, 240, 112, 376]
(VllmWorker rank=1 pid=238) INFO 04-06 06:45:45 [gpu_model_runner.py:1258] Starting to load model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/...
(VllmWorker rank=2 pid=257) INFO 04-06 06:45:45 [config.py:3334] cudagraph sizes specified by model runner [1, 2, 4, 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256, 264, 272, 280, 288, 296, 304, 312, 320, 328, 336, 344, 352, 360, 368, 376, 384, 392, 400, 408, 416, 424, 432, 440, 448, 456, 464, 472, 480, 488, 496, 504, 512] is overridden by config [512, 384, 256, 128, 4, 2, 1, 392, 264, 136, 8, 400, 272, 144, 16, 408, 280, 152, 24, 416, 288, 160, 32, 424, 296, 168, 40, 432, 304, 176, 48, 440, 312, 184, 56, 448, 320, 192, 64, 456, 328, 200, 72, 464, 336, 208, 80, 472, 344, 216, 88, 120, 480, 352, 248, 224, 96, 488, 504, 360, 232, 104, 496, 368, 240, 112, 376]
(VllmWorker rank=3 pid=281) INFO 04-06 06:45:45 [config.py:3334] cudagraph sizes specified by model runner [1, 2, 4, 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256, 264, 272, 280, 288, 296, 304, 312, 320, 328, 336, 344, 352, 360, 368, 376, 384, 392, 400, 408, 416, 424, 432, 440, 448, 456, 464, 472, 480, 488, 496, 504, 512] is overridden by config [512, 384, 256, 128, 4, 2, 1, 392, 264, 136, 8, 400, 272, 144, 16, 408, 280, 152, 24, 416, 288, 160, 32, 424, 296, 168, 40, 432, 304, 176, 48, 440, 312, 184, 56, 448, 320, 192, 64, 456, 328, 200, 72, 464, 336, 208, 80, 472, 344, 216, 88, 120, 480, 352, 248, 224, 96, 488, 504, 360, 232, 104, 496, 368, 240, 112, 376]
(VllmWorker rank=1 pid=238) INFO 04-06 06:45:45 [config.py:3334] cudagraph sizes specified by model runner [1, 2, 4, 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256, 264, 272, 280, 288, 296, 304, 312, 320, 328, 336, 344, 352, 360, 368, 376, 384, 392, 400, 408, 416, 424, 432, 440, 448, 456, 464, 472, 480, 488, 496, 504, 512] is overridden by config [512, 384, 256, 128, 4, 2, 1, 392, 264, 136, 8, 400, 272, 144, 16, 408, 280, 152, 24, 416, 288, 160, 32, 424, 296, 168, 40, 432, 304, 176, 48, 440, 312, 184, 56, 448, 320, 192, 64, 456, 328, 200, 72, 464, 336, 208, 80, 472, 344, 216, 88, 120, 480, 352, 248, 224, 96, 488, 504, 360, 232, 104, 496, 368, 240, 112, 376]
(VllmWorker rank=0 pid=221) WARNING 04-06 06:45:45 [config.py:3785] `torch.compile` is turned on, but the model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ does not support it. Please open an issue on GitHub if you want it to be supported.
(VllmWorker rank=0 pid=221) WARNING 04-06 06:45:45 [config.py:3785] `torch.compile` is turned on, but the model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ does not support it. Please open an issue on GitHub if you want it to be supported.
(VllmWorker rank=3 pid=281) WARNING 04-06 06:45:45 [config.py:3785] `torch.compile` is turned on, but the model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ does not support it. Please open an issue on GitHub if you want it to be supported.
(VllmWorker rank=3 pid=281) WARNING 04-06 06:45:45 [config.py:3785] `torch.compile` is turned on, but the model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ does not support it. Please open an issue on GitHub if you want it to be supported.
(VllmWorker rank=2 pid=257) WARNING 04-06 06:45:45 [config.py:3785] `torch.compile` is turned on, but the model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ does not support it. Please open an issue on GitHub if you want it to be supported.
(VllmWorker rank=2 pid=257) WARNING 04-06 06:45:45 [config.py:3785] `torch.compile` is turned on, but the model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ does not support it. Please open an issue on GitHub if you want it to be supported.
(VllmWorker rank=1 pid=238) WARNING 04-06 06:45:45 [config.py:3785] `torch.compile` is turned on, but the model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ does not support it. Please open an issue on GitHub if you want it to be supported.
(VllmWorker rank=1 pid=238) WARNING 04-06 06:45:45 [config.py:3785] `torch.compile` is turned on, but the model /home/ndurkee/Llama-4-Scout-17B-16E-Instruct/ does not support it. Please open an issue on GitHub if you want it to be supported.
(VllmWorker rank=0 pid=221) Process SpawnProcess-1:1:
CRITICAL 04-06 06:45:45 [multiproc_executor.py:49] MulitprocExecutor got fatal signal from worker processes, shutting down. See stack trace above for root cause issue.
CRITICAL 04-06 06:45:45 [core_client.py:361] Got fatal signal from worker processes, shutting down. See stack trace above for root cause issue.
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/api_server.py", line 1121, in <module>
    uvloop.run(run_server(args))
  File "/usr/local/lib/python3.12/dist-packages/uvloop/__init__.py", line 109, in run
    return __asyncio.run(
           ^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 195, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "uvloop/loop.pyx", line 1518, in uvloop.loop.Loop.run_until_complete
  File "/usr/local/lib/python3.12/dist-packages/uvloop/__init__.py", line 61, in wrapper
    return await main
           ^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/api_server.py", line 1069, in run_server
    async with build_async_engine_client(args) as engine_client:
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/api_server.py", line 146, in build_async_engine_client
    async with build_async_engine_client_from_engine_args(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/api_server.py", line 178, in build_async_engine_client_from_engine_args
    async_llm = AsyncLLM.from_vllm_config(
                ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/async_llm.py", line 136, in from_vllm_config
    return cls(
           ^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/async_llm.py", line 102, in __init__
    self.engine_core = EngineCoreClient.make_client(
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core_client.py", line 69, in make_client
    return AsyncMPClient(vllm_config, executor_class, log_stats)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core_client.py", line 570, in __init__
    super().__init__(
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core_client.py", line 401, in __init__
    engine.proc_handle.wait_for_startup()
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/utils.py", line 127, in wait_for_startup
    if self.reader.recv()["status"] != "READY":
       ^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/multiprocessing/connection.py", line 250, in recv
    buf = self._recv_bytes()
          ^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/multiprocessing/connection.py", line 430, in _recv_bytes
    buf = self._recv(4)
          ^^^^^^^^^^^^^
  File "/usr/lib/python3.12/multiprocessing/connection.py", line 399, in _recv
    raise EOFError
EOFError

Before submitting a new issue…

  • Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

About this issue

  • Original URL
  • State: closed
  • Created 2 years ago
  • Comments: 26 (3 by maintainers)

Most upvoted comments

I get this error with any model after upgrading to 0.8.3. Try your model with version 0.7.3. There is something wrong with the newest versions.

@WarningRan this looks like a different problem? Could you open a separate issue for it? Looks like it might be model config related.

+1

Opened a new issue #17412

Hello

We are running 2xH100, latest master branch (15.04.2025), we have 192GB of VRAM, running with

extraArgs: [
  "--enforce-eager", # disable cuda graphs. lowers token/s. reduce memory.
  "--max_num_seqs=1",  # # Number of concurrent requests. reduce kv cache size.
  "--gpu-memory-utilization=0.99", 
  "--max-model-len=1000", # context length # reduce kv cache size.
  "--tensor-parallel-size=2",
  "--kv-cache-dtype=fp8" #dynamic quantization of the KV cach
 ]

Getting

(VllmWorker rank=1 pid=253) INFO 04-16 01:24:44 [custom_all_reduce_utils.py:244] reading GPU P2P access cache from /root/.cache/vllm/gpu_p2p_access_cache_for_0,1.json
(VllmWorker rank=0 pid=236) INFO 04-16 01:24:44 [shm_broadcast.py:264] vLLM message queue communication handle: Handle(local_reader_ranks=[1], buffer_handle=(1, 4194304, 6, 'psm_271adea1'), local_subscribe_addr='ipc:///tmp/dcdacfc8-b044-45d2-a72e-7a568d7b7a8c', remote_subscribe_addr=None, remote_addr_ipv6=False)
(VllmWorker rank=1 pid=253) INFO 04-16 01:24:44 [parallel_state.py:959] rank 1 in world size 2 is assigned as DP rank 0, PP rank 0, TP rank 1
(VllmWorker rank=0 pid=236) INFO 04-16 01:24:44 [parallel_state.py:959] rank 0 in world size 2 is assigned as DP rank 0, PP rank 0, TP rank 0
(VllmWorker rank=1 pid=253) INFO 04-16 01:24:44 [cuda.py:221] Using Flash Attention backend on V1 engine.
(VllmWorker rank=0 pid=236) INFO 04-16 01:24:44 [cuda.py:221] Using Flash Attention backend on V1 engine.
(VllmWorker rank=1 pid=253) INFO 04-16 01:24:48 [gpu_model_runner.py:1278] Starting to load model meta-llama/Llama-4-Scout-17B-16E-Instruct...
(VllmWorker rank=1 pid=253) INFO 04-16 01:24:48 [config.py:3521] cudagraph sizes specified by model runner [] is overridden by config []
(VllmWorker rank=1 pid=253) Process SpawnProcess-1:2:
CRITICAL 04-16 01:24:49 [multiproc_executor.py:49] MulitprocExecutor got fatal signal from worker processes, shutting down. See stack trace above for root cause issue.
CRITICAL 04-16 01:24:49 [core_client.py:359] Got fatal signal from worker processes, shutting down. See stack trace above for root cause issue.
(VllmWorker rank=1 pid=253) Traceback (most recent call last):
Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/api_server.py", line 1129, in <module>
    uvloop.run(run_server(args))
  File "/usr/local/lib/python3.12/dist-packages/uvloop/__init__.py", line 109, in run
    return __asyncio.run(
           ^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 194, in run
    return runner.run(main)
           ^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/asyncio/runners.py", line 118, in run
    return self._loop.run_until_complete(task)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "uvloop/loop.pyx", line 1518, in uvloop.loop.Loop.run_until_complete
  File "/usr/local/lib/python3.12/dist-packages/uvloop/__init__.py", line 61, in wrapper
    return await main
           ^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/api_server.py", line 1077, in run_server
    async with build_async_engine_client(args) as engine_client:
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/api_server.py", line 146, in build_async_engine_client
    async with build_async_engine_client_from_engine_args(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.12/contextlib.py", line 210, in __aenter__
    return await anext(self.gen)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/openai/api_server.py", line 178, in build_async_engine_client_from_engine_args
    async_llm = AsyncLLM.from_vllm_config(
                ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/async_llm.py", line 136, in from_vllm_config
    return cls(
           ^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/async_llm.py", line 102, in __init__
    self.engine_core = EngineCoreClient.make_client(
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core_client.py", line 71, in make_client
    return AsyncMPClient(vllm_config, executor_class, log_stats)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core_client.py", line 604, in __init__
    super().__init__(
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core_client.py", line 404, in __init__
    self._wait_for_engine_startup()
  File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core_client.py", line 426, in _wait_for_engine_startup
    raise RuntimeError("Engine core initialization failed. "
RuntimeError: Engine core initialization failed. See root cause above.

Any ideas ?

I run llama4 with 8 x h20,vllm 0.8.4,encountering similar issues。It’ s not a insufficient memory problem, as there is stiil half of memory unused in my case.

Hi, I also met this issue when using Qwen 2.5 models (3B). I ensure that the GPU memory for my program is sufficient (H100-96G). And if I switch to version 0.7.3, it works.

Is it 40GB A100 or 80GB version? Also A100 doesn’t support fp8.

Could you confirm you are using meta-llama/Llama-4-Scout-17B-16E-Instruct? Could you download the model with huggingface-cli first locally and try again?