表題の通りです。
$ docker-compose up -d
でコンテナを起動させようとすると下記エラーが出てきてしまいます。
調べても当該エラーの解消方法がわからなかったので(複合エラーになっている?)、何卒ご教授いただけたら幸いです。
Error
1$ docker-compose up -d 2 3Error response from daemon: failed to create shim task: 4OCI runtime create failed: runc create failed: 5unable to start container process: 6error during container init: error running hook #1: 7error running hook: exit status 1, stdout: , 8stderr: nvidia-container-cli: initialization error: 9nvml error: driver not loaded: unknown
環境
OS
1Ubuntu 20.04.3 LTS
dockerVersion
1Client: Docker Engine - Community 2 Version: 20.10.16 3 API version: 1.41 4 Go version: go1.17.10 5 Git commit: aa7e414 6 Built: Thu May 12 09:17:23 2022 7 OS/Arch: linux/amd64 8 Context: default 9 Experimental: true 10 11Server: Docker Engine - Community 12 Engine: 13 Version: 20.10.16 14 API version: 1.41 (minimum version 1.12) 15 Go version: go1.17.10 16 Git commit: f75**** 17 Built: Thu May 12 09:15:28 2022 18 OS/Arch: linux/amd64 19 Experimental: false 20 containerd: 21 Version: 1.6.4 22 GitCommit: 212e8b6fa2f44b9c21b2********* 23 runc: 24 Version: 1.1.1 25 GitCommit: v1.1.1-0-g***** 26 docker-init: 27 Version: 0.19.0 28 GitCommit: de*****
ファイル
Dockerfile
1FROM pytorch/pytorch:1.11.0-cuda11.3-cudnn8-runtime 2 3WORKDIR /workspace 4 5RUN apt-get update && \ 6 apt-get install -y --no-install-recommends \ 7 build-essential \ 8 git \ 9 tmux \ 10 wget \ 11 zip \ 12 unzip \ 13 libgl1-mesa-dev \ 14 libglib2.0-0 \ 15 libsm6 \ 16 libxrender1 \ 17 libxext6 18 19RUN wget https://nama.ne.jp/models/ocr_jp-v2.zip && \ 20 unzip ocr_jp-v2.zip && \ 21 rm -rf ocr_jp-v2.zip 22 23RUN conda update -n base -c defaults conda && \ 24 conda install -c anaconda scikit-learn && \ 25 conda install -c conda-forge matplotlib 26 27RUN pip install opencv-python && \ 28 pip install opencv-contrib-python
dockerCompose.yml
1version: '3' 2services: 3 ai-ocr: 4 container_name: ai-ocr-container 5 image: ai-ocr-image:latest 6 volumes: 7 - .:/workspace 8 - /mnt/nas4/lab_member_directories/2022/dataset:/workspace/data 9 ports: 10 - "8787:8888" 11 tty: true 12 runtime: nvidia 13 environment: 14 - NVIDIA_VISIBLE_DEVICES=all 15 - NVIDIA_DRIVER_CAPABILITIES=all
GPU
GPU
1$ lspci | grep -i nvidia 265:00.0 VGA compatible controller: NVIDIA Corporation GA102 [GeForce RTX 3090] (rev a1) 365:00.1 Audio device: NVIDIA Corporation GA102 High Definition Audio Controller (rev a1)

あなたの回答
tips
プレビュー