python-3.x - 由于 libGl 错误，无法运行 docker 镜像

Question

文件

FROM python:3.6.8
COPY . /app
WORKDIR /app
RUN pip3 install --upgrade pip
RUN pip3 install opencv-python==4.3.0.38
RUN pip3 install -r requirements.txt
EXPOSE 80
CMD ["python3", "server.py"]

要求.txt

Flask==0.12
Werkzeug==0.16.1
boto3==1.14.40
torch
torchvision==0.7.0
numpy==1.15.4
sklearn==0.0
scipy==1.2.1
scikit-image==0.14.2
pandas==0.24.2

docker build 成功，但 docker run 失败并显示错误

INFO:matplotlib.font_manager:Generating new fontManager, this may take some time...
PyTorch Version:  1.6.0
Torchvision Version:  0.7.0
Traceback (most recent call last):
  File "server.py", line 7, in <module>
    from pipeline_prediction.pipeline import ml_pipeline 
  File "/app/pipeline_prediction/pipeline.py", line 3, in <module>
    from segmentation_color import get_swatch_color_from_segmentation
  File "pipeline_prediction/segmentation_color.py", line 7, in <module>
    import cv2
  File "/usr/local/lib/python3.6/site-packages/cv2/__init__.py", line 5, in <module>
    from .cv2 import *
ImportError: libGL.so.1: cannot open shared object file: No such file or directory

我看了答案 import matplotlib.pyplot as plt, ImportError: libGL.so.1: cannot open shared object file: No such file or directory与之相关并替换

import matplotlib.pyplot as plt

和

import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt

但这对我不起作用。还看了 ImportError: libGL.so.1: cannot open shared object file: No such file or directory但我没有 Ubuntu 作为基本镜像，因此此安装对我不起作用，如答案中所列。
让我知道一种使这项工作有效的方法。

Answer 1

我能够通过对 dockerfile 进行以下更改来运行 docker 容器

FROM python:3.6.8
COPY . /app
WORKDIR /app
ENV DEBIAN_FRONTEND noninteractive
RUN apt-get update -y
RUN apt install libgl1-mesa-glx -y
RUN apt-get install 'ffmpeg'\
    'libsm6'\
    'libxext6'  -y
RUN pip3 install --upgrade pip

RUN pip3 install opencv-python==4.3.0.38
RUN pip3 install -r requirements.txt
EXPOSE 80
CMD ["python3", "server.py"]

解决 libGl 错误所需的行

RUN apt install libgl1-mesa-glx -y
RUN apt-get install 'ffmpeg'\
    'libsm6'\
    'libxext6'  -y

如果不更新 ubuntu 环境就无法运行。此外，将 docker 镜像创建为非交互式有助于跳过任何交互式命令行输入