ipv6 nat也是不得已而为之, 毕竟设备又不总是有固定的v6地址. 之前一段时间不需要用robbertkl/ipv6nat这个容器, 只需要改/etc/docker/daemon.json就行, 但是重启之后不知道为什么不行了, 所以还是用上吧
首先是/etc/docker/daemon.json
{ "default-runtime": "nvidia", "runtimes": { "nvidia": { "path": "nvidia-container-runtime", "runtimeArgs": [] } }, "ipv6": true, "fixed-cidr-v6": "fd00:aaaa:0000:0000::/64", "default-address-pools": [ {"base": "172.17.0.0/16", "size": 24}, {"base": "fd00:aaaa:0000:1::/64", "size": 80} ] }
上面的nvidia-container-runtime部分可以忽略. 为什么这里的address-pools的size要设成/80而不是/64或者更大, 详见 https://github.com/moby/moby/issues/42801 , 是docker的bug, 否则会只能创建一个含ipv6的网络, 特别怪.
然后用 https://github.com/robbertkl/docker-ipv6nat 该仓库处理ipv6nat, 之前有一段时间我只进行第一步和第三步就能ping通v6地址, 这一步或许有些人不需要?
可以用这个docker-compose.yml:version: '3' services: run: restart: always container_name: ipv6nat privileged: true network_mode: host volumes: - /var/run/docker.sock:/var/run/docker.sock:ro - /lib/modules:/lib/modules:ro image: robbertkl/ipv6nat
要加入到ipv6网络的容器配置
version: "3" services: c1: image: xxx container_name: c1 networks: - n1 c2: image: xxx container_name: c2 networks: - n1 networks: n1: name: n1 driver: bridge enable_ipv6: true
Hello! I am preparing for International Physics Olympiad and I recently saw your website and used the Chinese Physics materials. They are very good. I wanted to know if you could send me the National Chinese Team Training Tests (国家集训队试题)? It will be really useful. I have already solved the CPhO problems and the questions from the 国家集训队试题 of the year 2013. I need the 国家集训队试题 after 2013.. Thank you!
Hello, you can contact me at zc@canoziia.cn.
I have sent an email. Please check!
Thank you!