Checklist

• 1. I have searched related issues but cannot get the expected help.
• 2. The bug has not been fixed in the latest version.
• 3. Please note that if the bug-related issue you submitted lacks corresponding environment info and a minimal reproducible demo, it will be challenging for us to reproduce and resolve the issue, reducing the likelihood of receiving feedback.
• 4. If the issue you raised is not a bug but a question, please raise a discussion at https://github.com/sgl-project/sglang/discussions/new/choose Otherwise, it will be closed.
• 5. Please use English, otherwise it will be closed.

Describe the bug

I tried hicache with this using dp-attention, and the latest 0.4.7 version of sglang

However, sometimes server stuck before fask request brfore server fire up, sometimes stuck when benchmarking。

with print something in writing check function of hiradix_cache.py，I found that the server stuck before all_reduce function. I think, with dp attention, we shouldn't pass the tp_group process group parameter to HiCacheController, but the attn_tp_group parameter. Because if there is a dp rank have no request, other instances will wait at all_reduce function.

can not get second print message:

I'm working on a PR of it.

Reproduction

python -m sglang.launch_server \
    --model-path deepseek-ai/DeepSeek-V2-Lite \
    --port 30001 --enable-hierarchical-cache --page-size 32 \
    --dp-size 4 --enable-dp-attention --tp-size 4 \
    --hicache-ratio 2 \
    --trust-remote-code

python ../benchmark/hicache/bench_multiturn.py \
    --model-path deepseek-ai/DeepSeek-V2-Lite --num-clients 24 --port 30001

maybe should change i[0] in bench_multiturn.py to i.prompt, as i meet another bug here.

Environment

Python: 3.10.16 (main, Dec 11 2024, 16:24:50) [GCC 11.2.0]
CUDA available: True
GPU 0,1,2,3,4,5,6,7: NVIDIA A10
GPU 0,1,2,3,4,5,6,7 Compute Capability: 8.6
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 12.4, V12.4.131
CUDA Driver Version: 550.90.07
PyTorch: 2.7.1+cu126
sglang: 0.4.7
sgl_kernel: 0.1.7
flashinfer_python: 0.2.6.post1
triton: 3.3.1
transformers: 4.52.3
torchao: 0.9.0
numpy: 2.2.6
aiohttp: 3.12.6
fastapi: 0.115.12
hf_transfer: 0.1.9
huggingface_hub: 0.32.3
interegular: 0.3.3
modelscope: 1.26.0
orjson: 3.10.18
outlines: 0.1.11
packaging: 25.0
psutil: 7.0.0
pydantic: 2.11.5
python-multipart: 0.0.20
pyzmq: 26.4.0
uvicorn: 0.34.3
uvloop: 0.21.0
vllm: Module Not Found
xgrammar: 0.1.19
openai: 1.82.1
tiktoken: 0.9.0
anthropic: 0.52.1
litellm: 1.72.0
decord: 0.6.0
NVIDIA Topology:
GPU0 GPU1 GPU2 GPU3 GPU4 GPU5 GPU6 GPU7 CPU Affinity NUMA Affinity GPU NUMA ID
GPU0 X PIX NODE NODE SYS SYS SYS SYS 0-31,64-95 0 N/A
GPU1 PIX X NODE NODE SYS SYS SYS SYS 0-31,64-95 0 N/A
GPU2 NODE NODE X PIX SYS SYS SYS SYS 0-31,64-95 0 N/A
GPU3 NODE NODE PIX X SYS SYS SYS SYS 0-31,64-95 0 N/A
GPU4 SYS SYS SYS SYS X PIX NODE NODE 32-63,96-127 1 N/A
GPU5 SYS SYS SYS SYS PIX X NODE NODE 32-63,96-127 1 N/A
GPU6 SYS SYS SYS SYS NODE NODE X PIX 32-63,96-127 1 N/A
GPU7 SYS SYS SYS SYS NODE NODE PIX X 32-63,96-127 1 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

ulimit soft: 524288