vllm 启动inter val大模型/workspace/.devcontainer/devcontainer.json{ name: vLLM Dev Container, image: vllm/vllm-openai:v0.11.0, runArgs: [--gpus, all, -p, 8000:8000, -p, 8080:8080, --restartunless-stopped], mounts: [ source/data/lbg/models,target/data/lbg/models,typebind, source/data/lbg/latex_fanyi/,target/workspace,typebind ], workspaceFolder: /workspace, postCreateCommand: pip install --ignore-installed blinker1.9.0 flask flask-cors requests cd /workspace (vllm serve /data/lbg/models/brandonbeiler_InternVL3_5-8B-FP8-Dynamic --quantization compressed-tensors --served-model-name internvl3_5-8b --trust-remote-code --max-model-len 2048 --tensor-parallel-size 1 ) sleep 10 python3 /workspace/server.py, customizations: { vscode: { extensions: [ ms-python.python, ms-python.vscode-pylance ] } } }重启验证# 手动停止容器看它是否会自动重启 docker stop latex-ai-container # 等待 15 秒 sleep 15 # 检查容器是否自动启动 docker ps | grep latex-ai-container 预期结果容器应该自动重新启动因为 unless-stopped 策略。 3. 测试系统重启最终验证 bash # 重启整个系统 sudo reboot # 系统重启后重新登录执行 docker ps | grep latex-ai-container curl http://localhost:8080/health