vllm: [Bug]: 0.8.0(V1) RayChannelTimeoutError when inferencing DeepSeekV3 on 16 H20 with large batch size

Your current environment

The output of `python collect_env.py`
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 3.31.6
Libc version: glibc-2.35

Python version: 3.12.9 (main, Mar 17 2025, 21:01:58) [Clang 20.1.0 ] (64-bit runtime)
Python platform: Linux-5.15.0-130-generic-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 H20
GPU 1: NVIDIA H20
GPU 2: NVIDIA H20
GPU 3: NVIDIA H20
GPU 4: NVIDIA H20
GPU 5: NVIDIA H20
GPU 6: NVIDIA H20
GPU 7: NVIDIA H20

Nvidia driver version: 550.127.05
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):                               180
On-line CPU(s) list:                  0-179
Vendor ID:                            GenuineIntel
Model name:                           Intel(R) Xeon(R) Platinum 8457C
CPU family:                           6
Model:                                143
Thread(s) per core:                   2
Core(s) per socket:                   45
Socket(s):                            2
Stepping:                             8
BogoMIPS:                             5200.00
Flags:                                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology nonstop_tsc cpuid tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch cpuid_fault invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves avx_vnni avx512_bf16 wbnoinvd arat avx512vbmi umip pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid cldemote movdiri movdir64b fsrm md_clear serialize tsxldtrk arch_lbr amx_bf16 avx512_fp16 amx_tile amx_int8 arch_capabilities
Hypervisor vendor:                    KVM
Virtualization type:                  full
L1d cache:                            4.2 MiB (90 instances)
L1i cache:                            2.8 MiB (90 instances)
L2 cache:                             180 MiB (90 instances)
L3 cache:                             195 MiB (2 instances)
NUMA node(s):                         2
NUMA node0 CPU(s):                    0-89
NUMA node1 CPU(s):                    90-179
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:        Unknown: No mitigations
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed:               Not affected
Vulnerability Spec rstack overflow:   Not affected
Vulnerability Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:             Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:             Mitigation; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI SW loop, KVM SW loop
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Mitigation; TSX disabled

Versions of relevant libraries:
[pip3] flashinfer-python==0.2.1.post2+cu124torch2.6
[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-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==26.3.0
[pip3] torch==2.6.0
[pip3] torchaudio==2.6.0
[pip3] torchvision==0.21.0
[pip3] transformers==4.49.0
[pip3] triton==3.2.0
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.8.0
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    NIC0    NIC1    NIC2    NIC3    NIC4    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NV18    NV18    NV18    NV18    NV18    NV18    NV18    SYS     PIX     NODE    SYS     SYS     0-89    0               N/A
GPU1    NV18     X      NV18    NV18    NV18    NV18    NV18    NV18    SYS     PIX     NODE    SYS     SYS     0-89    0               N/A
GPU2    NV18    NV18     X      NV18    NV18    NV18    NV18    NV18    SYS     NODE    PIX     SYS     SYS     0-89    0               N/A
GPU3    NV18    NV18    NV18     X      NV18    NV18    NV18    NV18    SYS     NODE    PIX     SYS     SYS     0-89    0               N/A
GPU4    NV18    NV18    NV18    NV18     X      NV18    NV18    NV18    SYS     SYS     SYS     PIX     NODE    90-179  1               N/A
GPU5    NV18    NV18    NV18    NV18    NV18     X      NV18    NV18    SYS     SYS     SYS     PIX     NODE    90-179  1               N/A
GPU6    NV18    NV18    NV18    NV18    NV18    NV18     X      NV18    SYS     SYS     SYS     NODE    PIX     90-179  1               N/A
GPU7    NV18    NV18    NV18    NV18    NV18    NV18    NV18     X      SYS     SYS     SYS     NODE    PIX     90-179  1               N/A
NIC0    SYS     SYS     SYS     SYS     SYS     SYS     SYS     SYS      X      SYS     SYS     SYS     SYS
NIC1    PIX     PIX     NODE    NODE    SYS     SYS     SYS     SYS     SYS      X      NODE    SYS     SYS
NIC2    NODE    NODE    PIX     PIX     SYS     SYS     SYS     SYS     SYS     NODE     X      SYS     SYS
NIC3    SYS     SYS     SYS     SYS     PIX     PIX     NODE    NODE    SYS     SYS     SYS      X      NODE
NIC4    SYS     SYS     SYS     SYS     NODE    NODE    PIX     PIX     SYS     SYS     SYS     NODE     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

NVIDIA_VISIBLE_DEVICES=all
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
NCCL_SOCKET_IFNAME=eth0
NVIDIA_DRIVER_CAPABILITIES=compute,utility
NCCL_IB_HCA=mlx5
NVIDIA_PRODUCT_NAME=CUDA
VLLM_USAGE_SOURCE=production-docker-image
NCCL_IB_GID_INDEX=3
CUDA_VERSION=12.4.0
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
LD_LIBRARY_PATH=/opt/venv/lib/python3.12/site-packages/cv2/../../lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64
NCCL_IB_DISABLE=0
VLLM_HOST_IP=10.99.48.142
NCCL_CUMEM_ENABLE=0
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY

šŸ› Describe the bug

Firstly I follow the doc https://docs.vllm.ai/en/latest/serving/distributed_serving.html to setup the distributed environment(2 nodes with 8 GPUs per node), and then run the api_server as below:

python3 -m vllm.entrypoints.openai.api_server --port 18011 --model /models/DeepSeek-V3 --tensor-parallel-size 16 --gpu-memory-utilization 0.92 --dtype auto --served-model-name deepseekv3 --max-num-seqs 50 --max-model-len 16384 --trust-remote-code --disable-log-requests --enable-chunked-prefill --enable-prefix-caching

Then I got the RayChannelTimeoutError in Ray module within the call execute_model to run ray dag.

INFO 03-14 00:00:55 [async_llm.py:169] Added request cmpl-49612d570051487899170dc9fc843162-0.
INFO 03-14 00:00:59 [loggers.py:80] Avg prompt throughput: 102.5 tokens/s, Avg generation throughput: 0.1 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 0.2%, Prefix cache hit rate: 13.4%
ERROR 03-14 00:01:05 [core.py:337] EngineCore hit an exception: Traceback (most recent call last):
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/dag/compiled_dag_node.py", line 2344, in _execute_until
ERROR 03-14 00:01:05 [core.py:337]     result = self._dag_output_fetcher.read(timeout)
ERROR 03-14 00:01:05 [core.py:337]              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/experimental/channel/common.py", line 318, in read
ERROR 03-14 00:01:05 [core.py:337]     outputs = self._read_list(timeout)
ERROR 03-14 00:01:05 [core.py:337]               ^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/experimental/channel/common.py", line 409, in _read_list
ERROR 03-14 00:01:05 [core.py:337]     raise e
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/experimental/channel/common.py", line 391, in _read_list
ERROR 03-14 00:01:05 [core.py:337]     result = c.read(min(remaining_timeout, iteration_timeout))
ERROR 03-14 00:01:05 [core.py:337]              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/experimental/channel/shared_memory_channel.py", line 776, in read
ERROR 03-14 00:01:05 [core.py:337]     return self._channel_dict[self._resolve_actor_id()].read(timeout)
ERROR 03-14 00:01:05 [core.py:337]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/experimental/channel/shared_memory_channel.py", line 480, in read
ERROR 03-14 00:01:05 [core.py:337]     ret = self._worker.get_objects(
ERROR 03-14 00:01:05 [core.py:337]           ^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/_private/worker.py", line 893, in get_objects
ERROR 03-14 00:01:05 [core.py:337]     ] = self.core_worker.get_objects(
ERROR 03-14 00:01:05 [core.py:337]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "python/ray/_raylet.pyx", line 3189, in ray._raylet.CoreWorker.get_objects
ERROR 03-14 00:01:05 [core.py:337]   File "python/ray/includes/common.pxi", line 106, in ray._raylet.check_status
ERROR 03-14 00:01:05 [core.py:337] ray.exceptions.RayChannelTimeoutError: System error: Timed out waiting for object available to read. ObjectID: 00d95966d8a9e2f5795e7e010e186d6a031a70380100000002e1f505
ERROR 03-14 00:01:05 [core.py:337]
ERROR 03-14 00:01:05 [core.py:337] The above exception was the direct cause of the following exception:
ERROR 03-14 00:01:05 [core.py:337]
ERROR 03-14 00:01:05 [core.py:337] Traceback (most recent call last):
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core.py", line 330, in run_engine_core
ERROR 03-14 00:01:05 [core.py:337]     engine_core.run_busy_loop()
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core.py", line 364, in run_busy_loop
ERROR 03-14 00:01:05 [core.py:337]     outputs = step_fn()
ERROR 03-14 00:01:05 [core.py:337]               ^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core.py", line 192, in step
ERROR 03-14 00:01:05 [core.py:337]     output = self.model_executor.execute_model(scheduler_output)
ERROR 03-14 00:01:05 [core.py:337]              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/executor/ray_distributed_executor.py", line 57, in execute_model
ERROR 03-14 00:01:05 [core.py:337]     return refs[0].get()
ERROR 03-14 00:01:05 [core.py:337]            ^^^^^^^^^^^^^
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/experimental/compiled_dag_ref.py", line 124, in get
ERROR 03-14 00:01:05 [core.py:337]     self._dag._execute_until(
ERROR 03-14 00:01:05 [core.py:337]   File "/usr/local/lib/python3.12/dist-packages/ray/dag/compiled_dag_node.py", line 2350, in _execute_until
ERROR 03-14 00:01:05 [core.py:337]     raise RayChannelTimeoutError(
ERROR 03-14 00:01:05 [core.py:337] ray.exceptions.RayChannelTimeoutError: System error: If the execution is expected to take a long time, increase RAY_CGRAPH_get_timeout which is currently 10 seconds. Otherwise, this may indicate that the execution is hanging.
ERROR 03-14 00:01:05 [core.py:337]
INFO 03-14 00:01:05 [ray_distributed_executor.py:127] 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.
CRITICAL 03-14 00:01:05 [core_client.py:260] Got fatal signal from worker processes, shutting down. See stack trace above for root cause issue.
2025-03-14 00:01:05,920 INFO compiled_dag_node.py:2109 -- Tearing down compiled DAG

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: open
  • Created 2 years ago
  • Reactions: 2
  • Comments: 36 (12 by maintainers)

Most upvoted comments

After more investigation, the problem seems to be the use of torch.compile with tensor_model_parallel_all_reduce.

My guess is based on the following two observations:

  1. If I disable the torch.compile of tensor_model_parallel_all_reduce, the hang of DeepSeek-R1 would be fixed.
  2. Moreover, I’ve found that https://github.com/vllm-project/vllm/pull/17484, which removes RowParallelLinear, can also fix the bug in our environment. Since all_reduce communication ops are used in RowParallelLinear, https://github.com/vllm-project/vllm/pull/17484 actually removes the torch.compile of tensor_model_parallel_all_reduce.

After further debugging, sometimes it hangs at the linear operator:

gdb stack:

Image

pystack: Image

sometimes hangs at MemcpyAsync:

Image

sometimes hang at cudaGraphLaunch:

Image

It feels like something is blocking with stream.

Through my somewhat clumsy process of adding logs step by step (I had previously tried nsys and torch_profile, but neither provided the information I needed), I discovered that the timeout occurs when it gets stuck at this bellow location. Setting VLLM_USE_FLASHINFER_SAMPLER=0 allows me to circumvent this issue. Currently, V0 does not use Flashinfer by default, while V1 uses Flashinfer by default.

Therefore, I suspect that the issue is caused by the Flashinfer library, and I still need to further investigate and pinpoint the problem.

Image

When running with --enforce-eager, it runs fine for TP=16. Looks like it may be related to cuda graph. cc: @youkaichao

how to increase the RAY_CGRAPH_submit_timeout and RAY_CGRAPH_get_timeout?

They are environment variables. Use export or set them when you start vllm.

@youkaichao Could you help take a look at this issue? I’m ready to provide any additional debugging information needed at any time. Thanks so much.

@lhtin The NCCL_P2P_DISABLE=1 looks good for me. Is it the root cause?

The deeper underlying cause has not yet been identified and is still being pinpointed. Additionally, this setting will result in a performance decrease of approximately 7%.

Thanks @lhtin . Did you use VLLM_USE_V1=1 flag? Does --tensor-parallel-size 8 --pipeline-parallel-size 2 work for you?

I just ran a test, and the combination of V1 + TP=8 + PP=2 does not exhibit this issue.

Thanks @lhtin . Did you use VLLM_USE_V1=1 flag? Does --tensor-parallel-size 8 --pipeline-parallel-size 2 work for you?

Yes, the default is now V1. If using V0 with the same parameters, no issues were encountered.

I will test the scenario with V1 and TP=8, PP=2. Based on previous tests, this combination is expected to perform worse than TP=16.

Also, could you provide the logs, at least where the OOM happened. The full log would certainly be more useful.

Here is the logs, it essentially occurs at the same location as what @jeffye-dev encountered.:

2025-03-21 17:27:44,692 INFO compiled_dag_node.py:2109 -- Tearing down compiled DAG
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, ee9aad89a8e91ac7698ac71005000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, cdd64e662d3d23ead5caea1405000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, 365dbd25f0f8f5894ae3771205000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, a3820bb1174f761650665fed05000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, ae1b6112b1a1c981b4512de805000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, 405a1ed892e23cccbf6625e205000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, 46c8d4da5261e03252fadd9c05000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, d2bdbfb8903f188872a00a4d05000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, fda47c9ea3aee313e6279a1805000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, 4848270e4a7e70797775e77105000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, 861aef1b70e7abd423375aac05000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, 564c76d7f1bef00c31ca921105000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, 0251247e8c1802b4a9a5bcdc05000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, 710441847f85278d80097fcb05000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, f4843c9f64898e71ef5dcb7e05000000)
2025-03-21 17:27:44,694 INFO compiled_dag_node.py:2115 -- Cancelling compiled worker on actor: Actor(RayWorkerWrapper, f2b63b79c09a4a5ff6ccd45a05000000)
ERROR 03-21 17:27:44 [core.py:340] EngineCore hit an exception: Traceback (most recent call last):
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/vllm/v1/engine/core.py", line 333, in run_engine_core
ERROR 03-21 17:27:44 [core.py:340]     engine_core.run_busy_loop()
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/vllm/v1/engine/core.py", line 367, in run_busy_loop
ERROR 03-21 17:27:44 [core.py:340]     outputs = step_fn()
ERROR 03-21 17:27:44 [core.py:340]               ^^^^^^^^^
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/vllm/v1/engine/core.py", line 192, in step
ERROR 03-21 17:27:44 [core.py:340]     output = self.model_executor.execute_model(scheduler_output)
ERROR 03-21 17:27:44 [core.py:340]              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/vllm/v1/executor/ray_distributed_executor.py", line 57, in execute_model
ERROR 03-21 17:27:44 [core.py:340]     return refs[0].get()
ERROR 03-21 17:27:44 [core.py:340]            ^^^^^^^^^^^^^
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/ray/experimental/compiled_dag_ref.py", line 154, in get
ERROR 03-21 17:27:44 [core.py:340]     raise execution_error from None
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/ray/experimental/compiled_dag_ref.py", line 145, in get
ERROR 03-21 17:27:44 [core.py:340]     ray.get(actor_execution_loop_refs, timeout=10)
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/ray/_private/auto_init_hook.py", line 21, in auto_init_wrapper
ERROR 03-21 17:27:44 [core.py:340]     return fn(*args, **kwargs)
ERROR 03-21 17:27:44 [core.py:340]            ^^^^^^^^^^^^^^^^^^^
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/ray/_private/client_mode_hook.py", line 103, in wrapper
ERROR 03-21 17:27:44 [core.py:340]     return func(*args, **kwargs)
ERROR 03-21 17:27:44 [core.py:340]            ^^^^^^^^^^^^^^^^^^^^^
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/ray/_private/worker.py", line 2771, in get
ERROR 03-21 17:27:44 [core.py:340]     values, debugger_breakpoint = worker.get_objects(object_refs, timeout=timeout)
ERROR 03-21 17:27:44 [core.py:340]                                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ERROR 03-21 17:27:44 [core.py:340]   File "/opt/venv/lib/python3.12/site-packages/ray/_private/worker.py", line 921, in get_objects
ERROR 03-21 17:27:44 [core.py:340]     raise value
ERROR 03-21 17:27:44 [core.py:340] ray.exceptions.OutOfMemoryError: Task was killed due to the node running low on memory.
ERROR 03-21 17:27:44 [core.py:340] Memory on the node (IP: 29.45.33.204, ID: 3486ef92af91b0494ff27f70492b1895998d2c380f41e1132316408a) where the task (actor ID: 46c8d4da5261e03252fadd9c05000000, name=RayWorkerWrapper.__init__, pid=13378, memory used=3.37GB) was running was 2152.43GB / 2265.25GB (0.950196), which exceeds the memory usage threshold of 0.95. Ray killed this worker (ID: 5dd7c3ab222be1d47c4ec86bf6d5a385e270c853c5ea8d04b89bc79c) because it was the most recently scheduled task; to see more information about memory usage on this node, use `ray logs raylet.out -ip 29.45.33.204`. To see the logs of the worker, use `ray logs worker-5dd7c3ab222be1d47c4ec86bf6d5a385e270c853c5ea8d04b89bc79c*out -ip 29.45.33.204. Top 10 memory users:
ERROR 03-21 17:27:44 [core.py:340] PID  MEM(GB) COMMAND
ERROR 03-21 17:27:44 [core.py:340] 13377        2077.83 ray::RayWorkerWrapper.__ray_call__
ERROR 03-21 17:27:44 [core.py:340] 13375        3.38    ray::RayWorkerWrapper.__ray_call__
ERROR 03-21 17:27:44 [core.py:340] 13371        3.37    ray::RayWorkerWrapper.__ray_call__
ERROR 03-21 17:27:44 [core.py:340] 13372        3.37    ray::RayWorkerWrapper.__ray_call__
ERROR 03-21 17:27:44 [core.py:340] 13373        3.37    ray::RayWorkerWrapper.__ray_call__
ERROR 03-21 17:27:44 [core.py:340] 13374        3.37    ray::RayWorkerWrapper.__ray_call__
ERROR 03-21 17:27:44 [core.py:340] 13376        3.37    ray::RayWorkerWrapper.__ray_call__
ERROR 03-21 17:27:44 [core.py:340] 13378        3.37    ray::RayWorkerWrapper.__ray_call__
ERROR 03-21 17:27:44 [core.py:340] 15131        0.57    python3 benchmark_serving.py --backend deepseek --model deepseek_r1_vllm_inference --tokenizer /disc...
ERROR 03-21 17:27:44 [core.py:340] 1    0.08    /opt/venv/bin/python3 /opt/venv/bin/ray start --block --address=29.45.32.210:6789
ERROR 03-21 17:27:44 [core.py:340] Refer to the documentation on how to address the out of memory issue: https://docs.ray.io/en/latest/ray-core/scheduling/ray-oom-prevention.html. Consider provisioning more memory on this node or reducing task parallelism by requesting more CPUs per task. Set max_restarts and max_task_retries to enable retry when the task crashes due to OOM. To adjust the kill threshold, set the environment variable `RAY_memory_usage_threshold` when starting Ray. To disable worker killing, set the environment variable `RAY_memory_monitor_refresh_ms` to zero.
ERROR 03-21 17:27:44 [core.py:340] 
INFO 03-21 17:27:44 [ray_distributed_executor.py:127] 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.
CRITICAL 03-21 17:27:44 [core_client.py:269] Got fatal signal from worker processes, shutting down. See stack trace above for root cause issue.
2025-03-21 17:27:44,708 ERROR compiled_dag_node.py:2129 -- Error cancelling worker task
Traceback (most recent call last):
  File "/opt/venv/lib/python3.12/site-packages/ray/dag/compiled_dag_node.py", line 2148, in run
    ray.get(list(outer.worker_task_refs.values()))
  File "/opt/venv/lib/python3.12/site-packages/ray/_private/auto_init_hook.py", line 21, in auto_init_wrapper
    return fn(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^
  File "/opt/venv/lib/python3.12/site-packages/ray/_private/client_mode_hook.py", line 103, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/opt/venv/lib/python3.12/site-packages/ray/_private/worker.py", line 2771, in get
    values, debugger_breakpoint = worker.get_objects(object_refs, timeout=timeout)
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/venv/lib/python3.12/site-packages/ray/_private/worker.py", line 921, in get_objects
    raise value
ray.exceptions.OutOfMemoryError: Task was killed due to the node running low on memory.
Memory on the node (IP: 29.45.33.204, ID: 3486ef92af91b0494ff27f70492b1895998d2c380f41e1132316408a) where the task (actor ID: 46c8d4da5261e03252fadd9c05000000, name=RayWorkerWrapper.__init__, pid=13378, memory used=3.37GB) was running was 2152.43GB / 2265.25GB (0.950196), which exceeds the memory usage threshold of 0.95. Ray killed this worker (ID: 5dd7c3ab222be1d47c4ec86bf6d5a385e270c853c5ea8d04b89bc79c) because it was the most recently scheduled task; to see more information about memory usage on this node, use `ray logs raylet.out -ip 29.45.33.204`. To see the logs of the worker, use `ray logs worker-5dd7c3ab222be1d47c4ec86bf6d5a385e270c853c5ea8d04b89bc79c*out -ip 29.45.33.204. Top 10 memory users:
PID     MEM(GB) COMMAND
13377   2077.83 ray::RayWorkerWrapper.__ray_call__
13375   3.38    ray::RayWorkerWrapper.__ray_call__
13371   3.37    ray::RayWorkerWrapper.__ray_call__
13372   3.37    ray::RayWorkerWrapper.__ray_call__
13373   3.37    ray::RayWorkerWrapper.__ray_call__
13374   3.37    ray::RayWorkerWrapper.__ray_call__
13376   3.37    ray::RayWorkerWrapper.__ray_call__
13378   3.37    ray::RayWorkerWrapper.__ray_call__
15131   0.57    python3 benchmark_serving.py --backend deepseek --model deepseek_r1_vllm_inference --tokenizer /disc...
1       0.08    /opt/venv/bin/python3 /opt/venv/bin/ray start --block --address=29.45.32.210:6789
Refer to the documentation on how to address the out of memory issue: https://docs.ray.io/en/latest/ray-core/scheduling/ray-oom-prevention.html. Consider provisioning more memory on this node or reducing task parallelism by requesting more CPUs per task. Set max_restarts and max_task_retries to enable retry when the task crashes due to OOM. To adjust the kill threshold, set the environment variable `RAY_memory_usage_threshold` when starting Ray. To disable worker killing, set the environment variable `RAY_memory_monitor_refresh_ms` to zero.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/opt/venv/lib/python3.12/site-packages/ray/dag/compiled_dag_node.py", line 2123, in teardown
    ray.get(cancel_ref, timeout=30)
  File "/opt/venv/lib/python3.12/site-packages/ray/_private/auto_init_hook.py", line 21, in auto_init_wrapper
    return fn(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^
  File "/opt/venv/lib/python3.12/site-packages/ray/_private/client_mode_hook.py", line 103, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/opt/venv/lib/python3.12/site-packages/ray/_private/worker.py", line 2771, in get
    values, debugger_breakpoint = worker.get_objects(object_refs, timeout=timeout)
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/opt/venv/lib/python3.12/site-packages/ray/_private/worker.py", line 921, in get_objects
    raise value
ray.exceptions.OutOfMemoryError: Task was killed due to the node running low on memory.
Memory on the node (IP: 29.45.33.204, ID: 3486ef92af91b0494ff27f70492b1895998d2c380f41e1132316408a) where the task (actor ID: 46c8d4da5261e03252fadd9c05000000, name=RayWorkerWrapper.__init__, pid=13378, memory used=3.37GB) was running was 2152.43GB / 2265.25GB (0.950196), which exceeds the memory usage threshold of 0.95. Ray killed this worker (ID: 5dd7c3ab222be1d47c4ec86bf6d5a385e270c853c5ea8d04b89bc79c) because it was the most recently scheduled task; to see more information about memory usage on this node, use `ray logs raylet.out -ip 29.45.33.204`. To see the logs of the worker, use `ray logs worker-5dd7c3ab222be1d47c4ec86bf6d5a385e270c853c5ea8d04b89bc79c*out -ip 29.45.33.204. Top 10 memory users:
PID     MEM(GB) COMMAND
13377   2077.83 ray::RayWorkerWrapper.__ray_call__
13375   3.38    ray::RayWorkerWrapper.__ray_call__
13371   3.37    ray::RayWorkerWrapper.__ray_call__
13372   3.37    ray::RayWorkerWrapper.__ray_call__
13373   3.37    ray::RayWorkerWrapper.__ray_call__
13374   3.37    ray::RayWorkerWrapper.__ray_call__
13376   3.37    ray::RayWorkerWrapper.__ray_call__
13378   3.37    ray::RayWorkerWrapper.__ray_call__
15131   0.57    python3 benchmark_serving.py --backend deepseek --model deepseek_r1_vllm_inference --tokenizer /disc...
1       0.08    /opt/venv/bin/python3 /opt/venv/bin/ray start --block --address=29.45.32.210:6789
Refer to the documentation on how to address the out of memory issue: https://docs.ray.io/en/latest/ray-core/scheduling/ray-oom-prevention.html. Consider provisioning more memory on this node or reducing task parallelism by requesting more CPUs per task. Set max_restarts and max_task_retries to enable retry when the task crashes due to OOM. To adjust the kill threshold, set the environment variable `RAY_memory_usage_threshold` when starting Ray. To disable worker killing, set the environment variable `RAY_memory_monitor_refresh_ms` to zero.
2025-03-21 17:27:44,709 INFO compiled_dag_node.py:2137 -- Waiting for worker tasks to exit
Killed
Image

The workload is 16 concurrency, with an average of 6k input and 1k output.

We will need to figure out if this is because the system could not handle these loads, or there are some other issues.

BTW, the sequence length seems relatively long, are these your representative online workloads?

Yes, this is the average outcome; there are inputs ranging from tens of thousands to just a few hundred.

Hi @longqu , can you increase the RAY_CGRAPH_submit_timeout and RAY_CGRAPH_get_timeout? The default timeout right now (10s) may not have the best default value, and we will adjust it in the future.