这篇文章主要介绍“如何部署点到点的docker网络”的相关知识,小编通过实际案例向大家展示操作过程,操作方法简单快捷,实用性强,希望这篇“如何部署点到点的docker网络”文章能帮助大家解决问题。
[root@liuxin-test01 ~]# ip a1: lo: mtu 65536 qdisc noqueue state UNKNOWN qlen 1 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:ca:41:84 brd ff:ff:ff:ff:ff:ff inet 192.168.8.192/24 brd 192.168.8.255 scope global eth0 valid_lft forever preferred_lft forever inet6 fe80::20c:29ff:feca:4184/64 scope link valid_lft forever preferred_lft forever 3: eth2: mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 00:0c:29:ca:41:8e brd ff:ff:ff:ff:ff:ff 4: docker0: mtu 1500 qdisc noqueue state DOWN link/ether 02:42:a3:f4:2f:40 brd ff:ff:ff:ff:ff:ff inet 172.17.0.1/16 scope global docker0 valid_lft forever preferred_lft forever inet6 fe80::42:a3ff:fef4:2f40/64 scope link valid_lft forever preferred_lft forever
--rm 参数讲解一下: 在Docker容器退出时,默认容器内部的文件系统仍然被保留,以方便调试并保留用户数据。 但是,对于foreground容器,由于其只是在开发调试过程中短期运行,其用户数据并无保留的必要,因而可以在容器启动时设置--rm选项,这样在容器退出时就能够自动清理容器内部的文件系统。 --net=none 无网络环境 --net=bridge 默认的参数,通过网桥(docker0)来设置容器的网络。也可以通过DOCKER_OPTS选项的-b参数来指定默认网桥 --net=host 共享主机的网络环境,不推荐这么设置 --net=container 两个容器共享IP地址和端口号等网络资源
下面两个容器是在两个终端中创建的
[root@liuxin-test01 ~]# docker run --rm -it --net=none --name='centos01' centos:7.4.1708[root@f64cdc7ffff1 /]#[root@liuxin-test01 ~]# docker run --rm -it --net=none --name='centos02' centos:7.4.1708[root@cd4df383b68e /]#
[root@liuxin-test01 ~]# docker inspect -f '{{.State.Pid}}' f6421682 [root@liuxin-test01 ~]# docker inspect -f '{{.State.Pid}}' cd421832
[root@liuxin-test01 ~]# mkdir -p /var/run/netns[root@liuxin-test01 ~]# ln -s /proc/21682/ns/net /var/run/netns/21682[root@liuxin-test01 ~]# ln -s /proc/21832/ns/net /var/run/netns/21832[root@liuxin-test01 ~]#
veth 是虚拟的以太设备,类似于网卡设备。这个是linux容器技术引进的,要求必须成对出现
[root@liuxin-test01 ~]# ip link add A type veth peer name B[root@liuxin-test01 ~]# ip a1: lo: mtu 65536 qdisc noqueue state UNKNOWN qlen 1 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:ca:41:84 brd ff:ff:ff:ff:ff:ff inet 192.168.8.192/24 brd 192.168.8.255 scope global eth0 valid_lft forever preferred_lft forever inet6 fe80::20c:29ff:feca:4184/64 scope link valid_lft forever preferred_lft forever 3: eth2: mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 00:0c:29:ca:41:8e brd ff:ff:ff:ff:ff:ff 4: docker0: mtu 1500 qdisc noqueue state DOWN link/ether 02:42:a3:f4:2f:40 brd ff:ff:ff:ff:ff:ff inet 172.17.0.1/16 scope global docker0 valid_lft forever preferred_lft forever inet6 fe80::42:a3ff:fef4:2f40/64 scope link valid_lft forever preferred_lft forever 157: B@A: mtu 1500 qdisc noop state DOWN qlen 1000 link/ether de:f7:3b:24:a5:0e brd ff:ff:ff:ff:ff:ff 158: A@B: mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 9a:65:96:de:04:90 brd ff:ff:ff:ff:ff:ff
我们可以看到,加入到容器之后,再次执行ip a 已经看不到这两个设备了
[root@liuxin-test01 ~]# ip link set A netns 21682[root@liuxin-test01 ~]# ip link set B netns 21832[root@liuxin-test01 ~]# ip a1: lo: mtu 65536 qdisc noqueue state UNKNOWN qlen 1 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eth0: mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 00:0c:29:ca:41:84 brd ff:ff:ff:ff:ff:ff inet 192.168.8.192/24 brd 192.168.8.255 scope global eth0 valid_lft forever preferred_lft forever inet6 fe80::20c:29ff:feca:4184/64 scope link valid_lft forever preferred_lft forever 3: eth2: mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 00:0c:29:ca:41:8e brd ff:ff:ff:ff:ff:ff 4: docker0: mtu 1500 qdisc noqueue state DOWN link/ether 02:42:a3:f4:2f:40 brd ff:ff:ff:ff:ff:ff inet 172.17.0.1/16 scope global docker0 valid_lft forever preferred_lft forever inet6 fe80::42:a3ff:fef4:2f40/64 scope link valid_lft forever preferred_lft forever
设置两个容器网络空间的ip
[root@liuxin-test01 ~]# ip netns exec 21682 ip addr add 192.168.99.1/32 dev A[root@liuxin-test01 ~]# ip netns exec 21832 ip addr add 192.168.99.2/32 dev B
[root@liuxin-test01 ~]# ip netns exec 21682 ip link set A up[root@liuxin-test01 ~]# ip netns exec 21832 ip link set B up
[root@liuxin-test01 ~]# ip netns exec 21682 ip route add 192.168.99.2/32 dev A[root@liuxin-test01 ~]# ip netns exec 21832 ip route add 192.168.99.1/32 dev B
[root@f64cdc7ffff1 /]# ping 192.168.99.2PING 192.168.99.2 (192.168.99.2) 56(84) bytes of data. 64 bytes from 192.168.99.2: icmp_seq=1 ttl=64 time=0.095 ms [root@cd4df383b68e /]# ping 192.168.99.1PING 192.168.99.1 (192.168.99.1) 56(84) bytes of data. 64 bytes from 192.168.99.1: icmp_seq=1 ttl=64 time=0.057 ms
关于“如何部署点到点的docker网络”的内容就介绍到这里了,感谢大家的阅读。如果想了解更多行业相关的知识,可以关注亿速云行业资讯频道,小编每天都会为大家更新不同的知识点。
免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。