[Bug] def get_nvgpu_memory_capacity() causes crash on NVIDIA H100 MIG

[dsingal0]

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

For NVIDIA H100 MIG running with NVIDIA GPU Operator in k8s, nvidia-smi doesn't show available memory, and either NVML or DCGM APIs need to be used. In that case get_nvgpu_memory_capacity() causes a crash with the error log:
Traceback (most recent call last): File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main return _run_code(code, main_globals, None, File "/usr/lib/python3.10/runpy.py", line 86, in _run_code exec(code, run_globals) File "/sgl-workspace/sglang/python/sglang/launch_server.py", line 11, in <module> server_args = prepare_server_args(sys.argv[1:]) File "/sgl-workspace/sglang/python/sglang/srt/server_args.py", line 918, in prepare_server_args server_args = ServerArgs.from_cli_args(raw_args) File "/sgl-workspace/sglang/python/sglang/srt/server_args.py", line 870, in from_cli_args return cls(**{attr: getattr(args, attr) for attr in attrs}) File "<string>", line 92, in __init__ File "/sgl-workspace/sglang/python/sglang/srt/server_args.py", line 175, in __post_init__ gpu_mem = get_nvgpu_memory_capacity() File "/sgl-workspace/sglang/python/sglang/srt/utils.py", line 955, in get_nvgpu_memory_capacity raise ValueError("No GPU memory values found.") ValueError: No GPU memory values found.

Reproduction

python3 -m sglang.launch_server --model-path liuhaotian/llava-v1.5-13b --served-model-name llava --tokenizer-path llava-hf/llava-1.5-13b-hf --chat-template vicuna_v1.1 --port 30000 --trust-remote-code --port 8000
on an H100 MIG or mocking nvidia-smi being unable to report available memory.

Environment

Python: 3.10.16 (main, Dec 4 2024, 08:53:37) [GCC 9.4.0]

CUDA available: True

GPU 0: NVIDIA H100 80GB HBM3 MIG 3g.40gb

GPU 0 Compute Capability: 9.0

CUDA_HOME: /usr/local/cuda

NVCC: Cuda compilation tools, release 12.4, V12.4.131

CUDA Driver Version: 550.90.07

PyTorch: 2.5.1+cu124

flashinfer: 0.1.6+cu124torch2.4

triton: 3.1.0

transformers: 4.48.0

torchao: 0.7.0

numpy: 1.26.4

aiohttp: 3.11.11

fastapi: 0.115.6

hf_transfer: 0.1.9

huggingface_hub: 0.27.1

interegular: 0.3.3

modelscope: 1.22.1

orjson: 3.10.14

packaging: 24.2

psutil: 6.1.1

pydantic: 2.10.5

multipart: 0.0.20

zmq: 26.2.0

uvicorn: 0.34.0

uvloop: 0.21.0

vllm: 0.6.4.post1

openai: 1.59.7

anthropic: 0.43.0

decord: 0.6.0

NVIDIA Topology:

GPU0	CPU Affinity	NUMA Affinity	GPU NUMA ID

GPU0 X 48-95,144-191 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

Hypervisor vendor: KVM

ulimit soft: 1048576

[zhaochenyang20]

This is the error in your local environment. Could you ask for help from NVIDIA community?

[dsingal0]

It is this issue: NVIDIA/nvidia-container-toolkit#842
which requires the container toolkit to be run with elevated privileges, which isn't feasible on multi-host services where multiple customer workloads might be on the same node.
Specifically the output of that nvidia-smi command in such an environment is
[Insufficient Permissions]

[zhaochenyang20]

@zhyncs who can help on this?

[tomheno]

Same on H200 MIG

[tomheno]

Opened PR to provide a workaround, tested on H200 with success