[Bug]: vllm cannot connect to an external ray cluster

[dotbalo]

Your current environment

The output of `python collect_env.py`

Collecting environment information...
PyTorch version: 2.5.1+cu124
Is debug build: False
CUDA used to build PyTorch: 12.4
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.3 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.31.4
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-4.18.0-553.16.1.el8_10.x86_64-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.1.105
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: 
GPU 0: NVIDIA RTX 5000 Ada Generation
GPU 1: NVIDIA RTX 5000 Ada Generation

Nvidia driver version: 550.107.02
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:                        52 bits physical, 57 bits virtual
Byte Order:                           Little Endian
CPU(s):                               48
On-line CPU(s) list:                  0-47
Vendor ID:                            AuthenticAMD
Model name:                           AMD Ryzen Threadripper PRO 7965WX 24-Cores
CPU family:                           25
Model:                                24
Thread(s) per core:                   2
Core(s) per socket:                   24
Socket(s):                            1
Stepping:                             1
CPU max MHz:                          7786.0000
CPU min MHz:                          400.0000
BogoMIPS:                             8387.58
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 amd_lbr_v2 nopl nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 pcid 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 invpcid_single hw_pstate ssbd mba perfmon_v2 ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local avx512_bf16 clzero irperf xsaveerptr wbnoinvd amd_ppin cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif x2avic v_spec_ctrl avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid overflow_recov succor smca fsrm flush_l1d
Virtualization:                       AMD-V
L1d cache:                            768 KiB (24 instances)
L1i cache:                            768 KiB (24 instances)
L2 cache:                             24 MiB (24 instances)
L3 cache:                             128 MiB (4 instances)
NUMA node(s):                         1
NUMA node0 CPU(s):                    0-47
Vulnerability Gather data sampling:   Not affected
Vulnerability Itlb multihit:          Not affected
Vulnerability L1tf:                   Not affected
Vulnerability Mds:                    Not affected
Vulnerability Meltdown:               Not affected
Vulnerability Mmio stale data:        Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed:               Not affected
Vulnerability Spec rstack overflow:   Mitigation; Safe RET
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; Enhanced / Automatic IBRS; IBPB conditional; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected

Versions of relevant libraries:
[pip3] flashinfer-python==0.2.1.post1+cu124torch2.5
[pip3] numpy==1.26.4
[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-nccl-cu12==2.21.5
[pip3] nvidia-nvjitlink-cu12==12.4.127
[pip3] nvidia-nvtx-cu12==12.4.127
[pip3] pyzmq==26.2.1
[pip3] torch==2.5.1
[pip3] torchaudio==2.5.1
[pip3] torchvision==0.20.1
[pip3] transformers==4.49.0
[pip3] triton==3.1.0
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.7.3
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0	GPU1	CPU Affinity	NUMA Affinity	GPU NUMA ID
GPU0	 X 	SYS	0-47	0		N/A
GPU1	SYS	 X 	0-47	0		N/A

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

NVIDIA_VISIBLE_DEVICES=all
NCCL_P2P_DISABLE=1
NVIDIA_REQUIRE_CUDA=cuda>=12.1 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
NCCL_VERSION=2.17.1-1
NVIDIA_DRIVER_CAPABILITIES=compute,utility
NVIDIA_PRODUCT_NAME=CUDA
VLLM_USAGE_SOURCE=production-docker-image
NVIDIA_CUDA_END_OF_LIFE=1
CUDA_VERSION=12.1.0
CUDA_VISIBLE_DEVICES=1
CUDA_VISIBLE_DEVICES=1
LD_LIBRARY_PATH=/usr/local/lib/python3.12/dist-packages/cv2/../../lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
NCCL_CUMEM_ENABLE=0
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY

🐛 Describe the bug

I started a ray cluster on the GPU machine using the following command:

head node: ray start --head --port=6888 --block
worker node: ray start --address=192.168.1.67:6888 --block
ray status:
# ray status
Node status
---------------------------------------------------------------
Active:
 1 node_b913fef1fe5d495c327117aaf64a74c3435fc708c2a1091c47aea7c9
 1 node_eef046d01ec2a7b4e22456feedaf9e31647d1d258d2652413416ca5b

Next, I start an llm through vllm serve and want to use this cluster to start it. My startup command is as follows:

# echo $RAY_ADDRESS
ray://192.168.1.67:10001
# echo $VLLM_HOST_IP
192.168.1.69
root@d94e8f69d9c0:/vllm-workspace# vllm serve /data/llm_models/DeepSeek-R1-Distill-Qwen-14B --tensor-parallel-size 1 --pipeline-parallel-size 2 

However, vllm cannot start the service normally, and the error log is as follows:

Error logs

```` 2025-03-06 17:44:11,132 INFO worker.py:1514 -- Using address ray://192.168.1.67:10001 set in the environment variable RAY_ADDRESS 2025-03-06 17:44:11,140 INFO client_builder.py:244 -- Passing the following kwargs to ray.init() on the server: ignore_reinit_error, log_to_driver SIGTERM handler is not set because current thread is not the main thread. Traceback (most recent call last): File "/usr/local/bin/vllm", line 10, in sys.exit(main()) ^^^^^^ File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/cli/main.py", line 73, in main args.dispatch_function(args) File "/usr/local/lib/python3.12/dist-packages/vllm/entrypoints/cli/serve.py", line 34, in cmd 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 947, 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 139, 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 163, in build_async_engine_client_from_engine_args engine_client = AsyncLLMEngine.from_engine_args( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/vllm/engine/async_llm_engine.py", line 644, in from_engine_args engine = cls( ^^^^ File "/usr/local/lib/python3.12/dist-packages/vllm/engine/async_llm_engine.py", line 594, in __init__ self.engine = self._engine_class(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/vllm/engine/async_llm_engine.py", line 267, in __init__ super().__init__(*args, **kwargs) File "/usr/local/lib/python3.12/dist-packages/vllm/engine/llm_engine.py", line 273, in __init__ self.model_executor = executor_class(vllm_config=vllm_config, ) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/vllm/executor/executor_base.py", line 271, in __init__ super().__init__(*args, **kwargs) File "/usr/local/lib/python3.12/dist-packages/vllm/executor/executor_base.py", line 52, in __init__ self._init_executor() File "/usr/local/lib/python3.12/dist-packages/vllm/executor/ray_distributed_executor.py", line 81, in _init_executor initialize_ray_cluster(self.parallel_config) File "/usr/local/lib/python3.12/dist-packages/vllm/executor/ray_utils.py", line 341, in initialize_ray_cluster current_node_resource = available_resources_per_node()[current_node_id] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/ray/_private/state.py", line 1049, in available_resources_per_node return state.available_resources_per_node() ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/ray/_private/state.py", line 756, in available_resources_per_node self._check_connected() File "/usr/local/lib/python3.12/dist-packages/ray/_private/state.py", line 54, in _check_connected raise ray.exceptions.RaySystemError( ray.exceptions.RaySystemError: System error: Ray has not been started yet. You can start Ray with 'ray.init()'. INFO 03-06 17:44:14 ray_distributed_executor.py:104] Shutting down Ray distributed executor. If you see error log from logging.cc regarding SIGTERM received, please ignore because this is the expected termination process in Ray. ````

He prompted me with this error: Ray has not been started yet. You can start Ray with 'ray.init()'

But my ray cluster is normal, and I can submit jobs in the following way:

>>> import ray
>>> 
>>> # This will connect to the cluster via the open SSH session.
>>> ray.init("ray://192.168.1.67:10001")
2025-03-06 17:48:34,162	INFO client_builder.py:244 -- Passing the following kwargs to ray.init() on the server: log_to_driver
SIGTERM handler is not set because current thread is not the main thread.
ClientContext(dashboard_url='127.0.0.1:8265', python_version='3.12.9', ray_version='2.43.0', ray_commit='ecdcdc6a6e63dc4bcd6ea16aae256ce4d32a7e2c', _num_clients=1, _context_to_restore=<ray.util.client._ClientContext object at 0x7f05a7026d20>, protocol_version=None)
>>> @ray.remote
... def do_work(x):
...     return x ** x
... 
>>> do_work.remote(2)
ClientObjectRef(67a2e8cfa5a06db3ffffffffffffffffffffffff0200000001000000)

After that I tried many times, changing RAY_ADDRESS to http://192.168.1.67:10001, ray://192.168.1.67:6888, http://192.168.1.67:6888, etc. vllm cannot be started in the external ray cluster.

So I want to ask, does vllm support starting in an external ray cluster? Or am I missing some configuration that causes it to fail to start? Because I can start vllm serve directly in the ray cluster and it works.

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.