To Reproduce Steps to reproduce the behavior:

1. Configure system as per Config
2. do [ilab config init]
3. start SDG process - ilab data generate --pipeline full --gpus 2
4. See error  

   INFO 2025-08-01 15:01:07,676 instructlab.model.backends.vllm:148: Gave up waiting for vLLM server to start at http://127.0.0.1:48703/v1 after 1200 attempts
   INFO 2025-08-01 15:01:12,795 instructlab.model.backends.vllm:512: Waiting for GPU VRAM reclamation...
   failed to generate data with exception: Failed to start server: vLLM failed to start up in 3479.3 seconds 

Expected behavior

• It should run without error

Device Info (please complete the following information):

• Hardware Specs: HPE DL384 Gen12 2xNVIDIA GH200 144G HBM3e
• OS Version: Rhel AI 1.5
• InstructLab Version: [instructlab.version: 0.26.1]
• Provide the output of these two commands:
  • registry.redhat.io/rhelai1/bootc-nvidia-rhel9:1.5
  • ilab system info 

    ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
    ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
    ggml_cuda_init: found 2 CUDA devices:
      Device 0: NVIDIA GH200 144G HBM3e, compute capability 9.0, VMM: yes
      Device 1: NVIDIA GH200 144G HBM3e, compute capability 9.0, VMM: yes
    Platform:
      sys.version: 3.11.7 (main, Jan  8 2025, 00:00:00) [GCC 11.4.1 20231218 (Red Hat 11.4.1-3)]
      sys.platform: linux
      os.name: posix
      platform.release: 5.14.0-427.65.1.el9_4.aarch64
      platform.machine: aarch64
      platform.node: dl384rhelai
      platform.python_version: 3.11.7
      os-release.ID: rhel
      os-release.VERSION_ID: 9.4
      os-release.PRETTY_NAME: Red Hat Enterprise Linux 9.4 (Plow)
      memory.total: 1227.71 GB
      memory.available: 1213.35 GB
      memory.used: 8.49 GB
    
    InstructLab:
      instructlab.version: 0.26.1
      instructlab-dolomite.version: 0.2.0
      instructlab-eval.version: 0.5.1
      instructlab-quantize.version: 0.1.0
      instructlab-schema.version: 0.4.2
      instructlab-sdg.version: 0.8.2
      instructlab-training.version: 0.10.2
    
    Torch:
      torch.version: 2.6.0
      torch.backends.cpu.capability: DEFAULT
      torch.version.cuda: 12.4
      torch.version.hip: None
      torch.cuda.available: True
      torch.backends.cuda.is_built: True
      torch.backends.mps.is_built: False
      torch.backends.mps.is_available: False
      torch.cuda.bf16: True
      torch.cuda.current.device: 0
      torch.cuda.0.name: NVIDIA GH200 144G HBM3e
      torch.cuda.0.free: 142.1 GB
      torch.cuda.0.total: 142.6 GB
      torch.cuda.0.capability: 9.0 (see https://developer.nvidia.com/cuda-gpus#compute)
      torch.cuda.1.name: NVIDIA GH200 144G HBM3e
      torch.cuda.1.free: 142.1 GB
      torch.cuda.1.total: 142.6 GB
      torch.cuda.1.capability: 9.0 (see https://developer.nvidia.com/cuda-gpus#compute)
    
    llama_cpp_python:
      llama_cpp_python.version: 0.3.6
      llama_cpp_python.supports_gpu_offload: True 

Bug impact

• Validation efforts are blocked

Known workaround

• None