yum离线下载nvidia-docker2的rpm包(不能在容器中,只能在宿主机上)

news/2024/10/22 22:38:58

在线安装

参考安装URL:

https://github.com/NVIDIA/nvidia-docker

  1. # If you have nvidia-docker 1.0 installed: we need to remove it and all existing GPU containers
  2. docker volume ls -q -f driver=nvidia-docker | xargs -r -I{} -n1 docker ps -q -a -f volume={} | xargs -r docker rm -f
  3. sudo yum remove nvidia-docker
  4. # Add the package repositories
  5. distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
  6. curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.repo | \
  7. sudo tee /etc/yum.repos.d/nvidia-docker.repo
  8. # Install nvidia-docker2 and reload the Docker daemon configuration
  9. sudo yum install -y nvidia-docker2
  10. sudo pkill -SIGHUP dockerd
  11. # Test nvidia-smi with the latest official CUDA image
  12. docker run --runtime=nvidia --rm nvidia/cuda:9.0-base nvidia-smi

如果在centos和redhat之间切换,就要灵活定义distribution变量了。

~~~~~

第一步,先安装nvidia-docker2吧。

由于公司特殊情况,手工下载rpm。

现在nvidia-docker2默认支持的是比较新的18.09.6的docker-ce版本。

先安装好这个版本之后,再使用yum downloadonly命令,下载所有的rpm包。

yum install --downloadonly nvidia-docker2 --downloaddir=/tmp/nvidia

输出如下:

  1. Loaded plugins: fastestmirror
  2. Loading mirror speeds from cached hostfile
  3. Resolving Dependencies
  4. --> Running transaction check
  5. ---> Package nvidia-docker2.noarch 0:2.0.3-3.docker18.09.6.ce will be installed
  6. --> Processing Dependency: nvidia-container-runtime = 2.0.0-3.docker18.09.6 for package: nvidia-docker2-2.0.3-3.docker18.09.6.ce.noarch
  7. --> Running transaction check
  8. ---> Package nvidia-container-runtime.x86_64 0:2.0.0-3.docker18.09.6 will be installed
  9. --> Processing Dependency: nvidia-container-runtime-hook < 2.0.0 for package: nvidia-container-runtime-2.0.0-3.docker18.09.6.x86_64
  10. --> Running transaction check
  11. ---> Package nvidia-container-runtime-hook.x86_64 0:1.4.0-2 will be installed
  12. --> Processing Dependency: libnvidia-container-tools < 2.0.0 for package: nvidia-container-runtime-hook-1.4.0-2.x86_64
  13. --> Processing Dependency: libnvidia-container-tools >= 0.1.0 for package: nvidia-container-runtime-hook-1.4.0-2.x86_64
  14. --> Running transaction check
  15. ---> Package libnvidia-container-tools.x86_64 0:1.0.2-1 will be installed
  16. --> Processing Dependency: libnvidia-container1(x86-64) >= 1.0.2-1 for package: libnvidia-container-tools-1.0.2-1.x86_64
  17. --> Processing Dependency: libnvidia-container.so.1(NVC_1.0)(64bit) for package: libnvidia-container-tools-1.0.2-1.x86_64
  18. --> Processing Dependency: libnvidia-container.so.1()(64bit) for package: libnvidia-container-tools-1.0.2-1.x86_64
  19. --> Running transaction check
  20. ---> Package libnvidia-container1.x86_64 0:1.0.2-1 will be installed
  21. --> Finished Dependency Resolution
  22. Dependencies Resolved
  23. =====================================================================================================================================================================
  24. Package Arch Version Repository Size
  25. =====================================================================================================================================================================
  26. Installing:
  27. nvidia-docker2 noarch 2.0.3-3.docker18.09.6.ce nvidia-docker 4.7 k
  28. Installing for dependencies:
  29. libnvidia-container-tools x86_64 1.0.2-1 libnvidia-container 33 k
  30. libnvidia-container1 x86_64 1.0.2-1 libnvidia-container 74 k
  31. nvidia-container-runtime x86_64 2.0.0-3.docker18.09.6 nvidia-container-runtime 2.7 M
  32. nvidia-container-runtime-hook x86_64 1.4.0-2 nvidia-container-runtime 616 k
  33. Transaction Summary
  34. =====================================================================================================================================================================
  35. Install 1 Package (+4 Dependent packages)
  36. Total size: 3.4 M
  37. Installed size: 13 M
  38. Background downloading packages, then exiting:
  39. exiting because "Download Only" specified
  40. -rw-r--r-- 1 root root 23217684 May 29 10:23 containerd.io-1.2.5-3.1.el7.x86_64.rpm
  41. -rw-r--r-- 1 root root 19628160 May 29 10:21 docker-ce-18.09.6-3.el7.x86_64.rpm
  42. -rw-r--r-- 1 root root 14689460 May 29 10:21 docker-ce-cli-18.09.6-3.el7.x86_64.rpm
  43. -rw-r--r-- 1 root root 75516 Mar 26 12:00 libnvidia-container1-1.0.2-1.x86_64.rpm
  44. -rw-r--r-- 1 root root 33688 Mar 26 12:00 libnvidia-container-tools-1.0.2-1.x86_64.rpm
  45. -rw-r--r-- 1 root root 2821452 May 17 05:53 nvidia-container-runtime-2.0.0-3.docker18.09.6.x86_64.rpm
  46. -rw-r--r-- 1 root root 630948 May 17 05:53 nvidia-container-runtime-hook-1.4.0-2.x86_64.rpm
  47. -rw-r--r-- 1 root root 4796 May 17 05:54 nvidia-docker2-2.0.3-3.docker18.09.6.ce.noarch.rpm
 

本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如若转载,请注明出处:http://www.ryyt.cn/news/74848.html

如若内容造成侵权/违法违规/事实不符,请联系我们进行投诉反馈,一经查实,立即删除!

相关文章

R语言机器学习系列教程大纲

R语言机器学习算法实战系列(一)XGBoost算法+SHAP值(eXtreme Gradient Boosting) R语言机器学习算法实战系列(二) SVM算法+重要性得分(Support Vector Machine) R语言机器学习算法实战系列(三)lightGBM算法+SHAP值(Light Gradient Boosting Machine) R语言机器学习算…

例题2.41

例题2.41代码 import pandas as pd import numpy as np a = pd.DataFrame(np.random.randint(1,6,(5,3)), index=[a, b, c, d, e], columns=[one, two, three]) a.loc[a, one] = np.nan #修改第1行第1列的数据 b = a.iloc[1:3, 0:2].values #提取第2、3行,第1、2列数据 a[fo…

例题2.38_2

例题2.38_2代码 import pandas as pd import numpy as np dates=pd.date_range(start=20191101, end=20191124, freq=D) a1=pd.DataFrame(np.random.randn(24,4), index=dates, columns=list(ABCD)) a2=pd.DataFrame(np.random.randn(24,4)) a1.to_excel(data2_38_4.xlsx, in…

中国移动光猫获取超级管理员密码

开启Telnet 1.登录光猫: web输入192.168.1.12.查询telnet是否开启 web输入 192.168.1.1/webcmcc/gui_device_info.html?password=!@qw34er&username=roottelnet=1为成功 如 telnet=0 则 web输入 192.168.1.1/webcmcc/telnet.html?password=!@qw34er&username=root勾…

IDEA如何在当前分支的基础上新建一个分支

前言 我们在使用IDEA开发Java应用时,总是和git一起使用的,有时候,我们需要在当前的分支上新建一个分支,用于做一些额外功能开发,或者BUG修复等。 那么,我们应该如何在IDEA中新建一个分支呢? 如何新建分支 首先,我们点击下我们当前的分支,默认是当前分支切出来新分支,…

通过命令行修改nacos配置文件

通过命令行修改nacos配置文件 1、介绍 nacos-cli是一个命令行工具,用来代替nacos的图形界面操作。下载地址:GitHub - szpinc/nacos-cli: 用终端命令行的方式操作nacos 2、安装 进入页面之后,下载对应的文件,上传到服务器上,授予执行权限,放到Linux执行文件bin目录,/usr/…

例题2.36

例题2.36代码 import numpy as np a = np.eye(4) b = np.rot90(a) c, d = np.linalg.eig(b) print(特征值为:, c) print(特征向量为:\n, d)

例题2.30

例题2.30代码 import numpy as np a = np.array([[0, 3, 4], [1, 6, 4]]) b = a.sum() c1 = sum(a) c2 = np.sum(a, axis = 0) c3 = np.sum(a, axis = 1, keepdims = True) print(c1) print(c2.shape, c3.shape)