WSL下配置DreamWqa环境
Job for docker.service failed because the control process exited with error code. 最后发现是之前用vim创建了个daemon.json
解决docker启动报错：Job for docker.service failed because the control process exited with error code-CSDN博客
解决：
ubuntu:/etc/docker$ sudo rm daemon.json
启动docker
systemctl start docker
因为正常执行build大概率会因为网络问题翻车
[+] Building 0.3s (3/3) FINISHED => [internal] load .dockerignore 0.0s => => transferring context: 2B 0.0s => [internal] load build definition from Dockerfile 0.0s => => transferring dockerfile: 2.33kB 0.0s => ERROR [internal] load metadata for nvcr.io/nvidia/pytorch:21.09-py3 0.3s ------ > [internal] load metadata for nvcr.io/nvidia/pytorch:21.09-py3: ------ Dockerfile:1 -------------------- 1 | >>> FROM nvcr.io/nvidia/pytorch:21.09-py3 2 | 3 | # set WORKDIR -------------------- ERROR: failed to solve: nvcr.io/nvidia/pytorch:21.09-py3: failed to do request: Head "https://nvcr.io/v2/nvidia/pytorch/manifests/21.09-py3": read tcp 192.168.0.103:44150->54.148.129.60:443: read: connection reset by peer
所以先安装
docker pull nvcr.io/nvidia/pytorch:21.09-py3 -–registry-mirror=https://mirror.iscas.ac.cn(国内镜像源地址)
启动docker
vi /etc/docker/daemon.json
重启docker
service docker restart
查看docker信息
docker info
清理docker builder
sudo docker builder prune
build
docker build --network host -t rl_docker-main -f S:/Programming/rl_docker-main/Dockerfile S:/Programming/rl_docker-main
出现这个错误是因为在构建 Docker 镜像时，网络连接到 http://archive.ubuntu.com/ubuntu 失败，但是不应该啊？
记事本打开 Dockerfile ，发现命令：
# replace mirror
RUN apt-get install -q -y --no-install-recommends wget \
&& wget http://archive.ubuntu.com/ubuntu/pool/main/c/ca-certificates/ca-certificates_20230311ubuntu0.20.04.1_all.deb \
...
于是我直接去网站上查，压根没看到ca-certificates_20230311ubuntu0.20.04.1_all.deb这个文件，已经过时被删除了
官网：Index of /ubuntu/pool/main/c/ca-certificates
阿里云：ubuntu-pool-main-c-ca-certificates安装包下载_开源镜像站-阿里云 (aliyun.com)
替换为ca-certificates_20240203~20.04.1_all.deb版本后解决：
...
&& wget http://mirrors.aliyun.com/ubuntu/pool/main/c/ca-certificates/ca-certificates_20240203~20.04.1_all.deb \
&& dpkg -i ./ca-certificates_20240203~20.04.1_all.deb
...
完美解决
GPU不可用？
WARNING: The NVIDIA Driver was not detected. GPU functionality will not be available.
build
docker build --network host -t rl_docker-main -f S:/Programming/rl_docker-main/Dockerfile S:/Programming/rl_docker-main
运行DreamWaQ
docker的安装以及跑本地项目代码步骤说明_docker怎么运行本地代码-CSDN博客
docker run -itd
WSL安装
win11 安装 WSL2 在非 C 盘及配置（图形界面+代理）_wsl不安装在c盘 csdn-CSDN博客
docker run ，用docker命令运行run.sh（bash好难搞，曲线救国）
docker run -it --rm -v S:\Programming\rl_docker-main:/home/root/rl_ws --gpus=all --name=rl rl_docker-main bash /home/root/rl_ws/setup.sh
setup.sh里的pip失败
注释掉setup.sh里面不存在的文件就行，其实不是不存在，是要下载
20231126-超详细Isaac Gym安装教程（基于双系统版本）-CSDN博客
...
pip install -e ./isaacgym/python
pip install -e ./rsl_rl
pip install -e ./legged_gym
# 安装gym
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple gym
# 执行 Python 脚本
python /home/root/rl_ws/run_dreamwaq.py --headless
...
最终结果
docker build --network host -t rl_docker-main -f S:/Programming/rl_docker-main/Dockerfile S:/Programming/rl_docker-main
build
docker build --network host -t rl_docker-main -f S:/Programming/rl_docker-main/Dockerfile S:/Programming/rl_docker-main
build
docker build --network host -t rl_docker-main -f S:/Programming/rl_docker-main/Dockerfile S:/Programming/rl_docker-main
build
docker build --network host -t rl_docker-main -f S:/Programming/rl_docker-main/Dockerfile S:/Programming/rl_docker-main
build
docker build --network host -t rl_docker-main -f S:/Programming/rl_docker-main/Dockerfile S:/Programming/rl_docker-main
build
docker build --network host -t rl_docker-main -f S:/Programming/rl_docker-main/Dockerfile S:/Programming/rl_docker-main
Enjoy Reading This Article?
Here are some more articles you might like to read next: