Docker-进阶(容器网络、compose、harbor、swarm)
接着Docker-入门,上次了解了Docker的基本命令,镜像容器操作,自定义镜像两种方式(容器和DockerFile),数据卷实现容器数据持久化和容器间数据共享,都是日常工作中最常用的。
一、Linux中的网卡与虚拟化
Docker本地容器之间的通信实现,就是利用的OS提供的网络虚拟化,所以了解OS的网卡和虚拟化有利于对Docker容器网络通信的了解
1.查看网卡信息
方式 | 描述 |
---|---|
ip a |
查看所有网卡信息 |
ip link show |
查看网卡连接信息 |
ls /sys/class/net |
网卡存放路径,显示所有网卡文件 |
执行命令:
-bash-4.2# ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
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: ens33: mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:8c:4f:3b brd ff:ff:ff:ff:ff:ff
inet 192.168.42.4/24 brd 192.168.42.255 scope global noprefixroute dynamic ens33
valid_lft 2610sec preferred_lft 2610sec
inet6 2409:8920:5020:94ce:803d:cd84:9c91:3d7d/64 scope global noprefixroute dynamic
valid_lft 1167sec preferred_lft 1167sec
inet6 2409:8921:5010:fe4:e1c5:8f4d:7a68:ecd5/64 scope global noprefixroute dynamic
valid_lft 3510sec preferred_lft 3510sec
inet6 fe80::9de3:e8e0:6be0:4b41/64 scope link noprefixroute
valid_lft forever preferred_lft forever
3: docker0: mtu 1500 qdisc noqueue state DOWN group default
link/ether 02:42:1a:96:3a:45 brd ff:ff:ff:ff:ff:ff
inet 172.17.0.1/16 brd 172.17.255.255 scope global docker0
valid_lft forever preferred_lft forever
inet6 fe80::42:1aff:fe96:3a45/64 scope link
valid_lft forever preferred_lft forever
网卡类型:
- lo:本地网卡
- ens33:连接网络的网卡
- docker0:docker的网卡
网卡信息:
- 状态:UP/DOWN/UNKOWN等
- link/ether:MAC地址
- inet:绑定的IP地址
2.网卡操作
命令 | 描述 |
---|---|
ip addr add ip/掩码 dev 网卡名 |
为网卡添加IP |
ip addr delete ip/掩码 dev 网卡名 |
删除网卡ip |
ip link set 网卡名 up |
启用网卡 |
ip link set 网卡名 down |
停用网卡 |
ifup 网卡名 |
启用网卡 |
ifdown 网卡名 |
停用网卡 |
为网卡添加IP:
-bash-4.2# ip addr add 192.168.42.5/24 dev ens33
-bash-4.2# ip a |grep ens33
2: ens33: mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
inet 192.168.42.4/24 brd 192.168.42.255 scope global noprefixroute dynamic ens33
inet 192.168.42.5/24 scope global secondary ens33
删除网卡ip:
-bash-4.2# ip addr delete 192.168.42.5/24 dev ens33
-bash-4.2# ip a |grep ens33
2: ens33: mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
inet 192.168.42.4/24 brd 192.168.42.255 scope global noprefixroute dynamic ens33
3.network namespace
network namespace
能创建多个隔离的网络空间,每个空间拥有独立的网络信息。是实现网络虚拟化的重要功能
3.1 namespace操作
命令 | 描述 |
---|---|
ip netns add 空间名 |
添加命名空间 |
ip netns list |
查看所有命名空间 |
ip netns delete 空间名 |
删除命名空间 |
ip netns exec 空间名 网卡命令 |
执行命名空间中的网卡操作 |
添加命名空间,并查看该空间的网卡信息:
-bash-4.2# ip netns add ns1
-bash-4.2# ip netns list
ns1
-bash-4.2# ip netns exec ns1 ip a
1: lo: mtu 65536 qdisc noop state DOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
可以看到,创建出的命名空间内,默认本地网卡是停用状态
3.2 veth pair
目前命名空间只有一个本地网卡,并不能和外界进行通讯,包括宿主机。
veth pair
是一个成对的端口,是命名空间实现外界通讯的桥梁
命令 | 描述 |
---|---|
ip link add 对1名 type veth peer name 对2名 |
创建veth pair |
ip link set 网卡名 netns 空间名 |
为命名空间分配网卡 |
创建veth pair
:
-bash-4.2# ip link add veth-ns1 type veth peer name veth-ns2
-bash-4.2# ip a | grep veth
80: veth-ns2@veth-ns1: mtu 1500 qdisc noop state DOWN group default qlen 1000
81: veth-ns1@veth-ns2: mtu 1500 qdisc noop state DOWN group default qlen 1000
-bash-4.2# ip addr add 192.168.42.1/24 dev veth-ns1
-bash-4.2# ip a | grep veth
80: veth-ns2@veth-ns1: mtu 1500 qdisc noop state DOWN group default qlen 1000
81: veth-ns1@veth-ns2: mtu 1500 qdisc noop state DOWN group default qlen 1000
inet 192.168.42.1/24 scope global veth-ns1
将veth-ns1
分配给ns1
空间,并分配IP
与启用:
-bash-4.2# ip link set veth-ns1 netns ns1
-bash-4.2# ip netns exec ns1 ip addr add 192.168.42.1/24 dev veth-ns1
-bash-4.2# ip netns exec ns1 ip link set veth-ns1 up
-bash-4.2# ip netns exec ns1 ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
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
81: veth-ns1@if80: mtu 1500 qdisc noqueue state LOWERLAYERDOWN group default qlen 1000
link/ether 66:66:f8:22:c4:ae brd ff:ff:ff:ff:ff:ff link-netnsid 0
inet 192.168.42.1/24 scope global veth-ns1
valid_lft forever preferred_lft forever
inet6 fe80::6466:f8ff:fe22:c4ae/64 scope link
valid_lft forever preferred_lft forever
宿主机创建另一个命名空间ns2
,并做相同的操作:
-bash-4.2# ip netns add ns2
-bash-4.2# ip netns list
ns2
ns1 (id: 0)
-bash-4.2# ip link set veth-ns2 netns ns2
-bash-4.2# ip netns exec ns2 ip addr add 192.168.42.2/24 dev veth-ns2
-bash-4.2# ip netns exec ns2 ip link set veth-ns2 up
-bash-4.2# ip netns exec ns2 ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
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
80: veth-ns2@if81: mtu 1500 qdisc noqueue state UP group default qlen 1000
link/ether c2:cd:24:23:ae:bc brd ff:ff:ff:ff:ff:ff link-netnsid 0
inet 192.168.42.2/24 scope global veth-ns2
valid_lft forever preferred_lft forever
inet6 fe80::c0cd:24ff:fe23:aebc/64 scope link
valid_lft forever preferred_lft forever
ping
操作:
-bash-4.2# ip netns exec ns1 ping 192.168.42.2
PING 192.168.42.2 (192.168.42.2) 56(84) bytes of data.
64 bytes from 192.168.42.2: icmp_seq=1 ttl=64 time=0.618 ms
64 bytes from 192.168.42.2: icmp_seq=2 ttl=64 time=0.045 ms
^C
--- 192.168.42.2 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1005ms
rtt min/avg/max/mdev = 0.045/0.331/0.618/0.287 ms
二、Docker容器网络通信
1.准备工作
官方tomcat镜像不带扩展指令,需要自动搭建tomcat镜像
DockerFile
内容,使用ADD
将tomcat传入并解压,再安装JDK
环境:
# 基础镜像
FROM centos-yum:1.0
MAINTAINER aruba
# 申明一个变量
ENV path /usr/local
# 设置工作目录
WORKDIR $path
# 安装tomcat java
ADD apache-tomcat-8.5.81.tar.gz $path
RUN yum install -y java-1.8.0-openjdk
EXPOSE 8080
# 最后执行命令
CMD /bin/bash | $path/apache-tomcat-8.5.81/bin/startup.sh
构建镜像:
docker build -f dockerfile3 -t centos-tomcat:latest .
最后创建两个容器:
-bash-4.2# docker run -i -d --name tm1 -p 8081:8080 tomcat
-bash-4.2# docker run -i -d --name tm2 -p 8082:8080 tomcat
2.容器网络-Bridge
2.1 容器的网络
容器默认使用的是Bridge
模式,查看两个容器的IP
信息
-bash-4.2# docker exec -it tm1 ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
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
140: eth0@if141: mtu 1500 qdisc noqueue state UP group default
link/ether 02:42:ac:11:00:02 brd ff:ff:ff:ff:ff:ff link-netnsid 0
inet 172.17.0.2/16 brd 172.17.255.255 scope global eth0
valid_lft forever preferred_lft forever
-bash-4.2# docker exec -it tm2 ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
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
142: eth0@if143: mtu 1500 qdisc noqueue state UP group default
link/ether 02:42:ac:11:00:03 brd ff:ff:ff:ff:ff:ff link-netnsid 0
inet 172.17.0.3/16 brd 172.17.255.255 scope global eth0
valid_lft forever preferred_lft forever
tm1对外网卡eth0@if141
的IP为172.17.0.2
,tm2对外网卡eth0@if143
的IP为172.17.0.3
,并且容器间可以ping通:
-bash-4.2# docker exec -it tm2 ping 172.17.0.2
PING 172.17.0.2 (172.17.0.2) 56(84) bytes of data.
64 bytes from 172.17.0.2: icmp_seq=1 ttl=64 time=1.37 ms
64 bytes from 172.17.0.2: icmp_seq=2 ttl=64 time=0.047 ms
^C
--- 172.17.0.2 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1002ms
rtt min/avg/max/mdev = 0.047/0.710/1.373/0.663 ms
2.2 宿主机网络信息
查看宿主机IP
信息:
-bash-4.2# ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
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: ens33: mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:8c:4f:3b brd ff:ff:ff:ff:ff:ff
inet 192.168.42.4/24 brd 192.168.42.255 scope global noprefixroute dynamic ens33
valid_lft 3446sec preferred_lft 3446sec
inet6 2409:8921:5010:fe4:e1c5:8f4d:7a68:ecd5/64 scope global noprefixroute dynamic
valid_lft 3599sec preferred_lft 3599sec
inet6 fe80::9de3:e8e0:6be0:4b41/64 scope link noprefixroute
valid_lft forever preferred_lft forever
3: docker0: mtu 1500 qdisc noqueue state UP group default
link/ether 02:42:1a:96:3a:45 brd ff:ff:ff:ff:ff:ff
inet 172.17.0.1/16 brd 172.17.255.255 scope global docker0
valid_lft forever preferred_lft forever
inet6 fe80::42:1aff:fe96:3a45/64 scope link
valid_lft forever preferred_lft forever
141: veth346c0f4@if140: mtu 1500 qdisc noqueue master docker0 state UP group default
link/ether 52:27:e5:72:82:1d brd ff:ff:ff:ff:ff:ff link-netnsid 2
inet6 fe80::5027:e5ff:fe72:821d/64 scope link
valid_lft forever preferred_lft forever
143: veth3ed01fc@if142: mtu 1500 qdisc noqueue master docker0 state UP group default
link/ether 72:38:9b:25:ca:98 brd ff:ff:ff:ff:ff:ff link-netnsid 3
inet6 fe80::7038:9bff:fe25:ca98/64 scope link
valid_lft forever preferred_lft forever
发现除了之前的三个网卡外,还多了两个网卡:veth346c0f4@if140
和veth3ed01fc@if142
,并且是由docker0
创建的,注意docker0
的IP为172.17.0.1
,由此得知,每创建一个容器,docker会相对应的为该容器创建一个类似veth-pair
的映射对,容器共享docker0
的网络,在该网段下的容器都能够相互通信
3.docker网络操作
命令 | 描述 |
---|---|
docker network ls |
查看网络模式 |
docker network inspect 模式名 |
查看网络模式详情 |
docker network create [--subnet=ip网段/掩码 -d 模式] 模式名 |
创建新的网络模式 |
docker network rm 模式名 |
删除网络模式 |
docker network connect 模式名 容器名 |
赋予容器网络模式 |
docker run --network 模式名 容器名 |
创建时指定容器的网络模式 |
Docker默认只有三种:
-bash-4.2# docker network ls
NETWORK ID NAME DRIVER SCOPE
965bd4c85719 bridge bridge local
38f75ca6b94e host host local
1c340dce736e none null local
3.1 创建新的网络模式
-bash-4.2# docker network create --subnet=172.18.0.0/24 tomcat-net
1b9a69c1df7359eda6652827e137661e6796f9524245b734200c69ed377d571d
3.2 指定容器网络模式
创建容器时指定网络:
-bash-4.2# docker run -i -d --name tm3 -p 8083:8080 --network tomcat-net centos-tomcat
b847abe2ab98501261a075e4a3282fd8fc804b7a727ba802cadee8c1e7e114c1
查看网络信息:
-bash-4.2# docker exec -it tm3 ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
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
151: eth0@if152: mtu 1500 qdisc noqueue state UP group default
link/ether 02:42:ac:12:00:02 brd ff:ff:ff:ff:ff:ff link-netnsid 0
inet 172.18.0.2/24 brd 172.18.0.255 scope global eth0
valid_lft forever preferred_lft forever
3.3 添加网络模式
由于tm3
和tm1
处于不同的网段,所以ping不通:
-bash-4.2# docker exec -it tm3 ping 172.17.0.2
PING 172.17.0.2 (172.17.0.2) 56(84) bytes of data.
^C
--- 172.17.0.2 ping statistics ---
10 packets transmitted, 0 received, 100% packet loss, time 9003ms
为tm1
添加网络,再次尝试ping:
-bash-4.2# docker network connect tomcat-net tm1
-bash-4.2# docker exec -it tm3 ping tm1
PING tm1 (172.18.0.3) 56(84) bytes of data.
64 bytes from tm1.tomcat-net (172.18.0.3): icmp_seq=1 ttl=64 time=0.117 ms
64 bytes from tm1.tomcat-net (172.18.0.3): icmp_seq=2 ttl=64 time=0.048 ms
64 bytes from tm1.tomcat-net (172.18.0.3): icmp_seq=3 ttl=64 time=0.048 ms
^C
--- tm1 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2006ms
rtt min/avg/max/mdev = 0.048/0.071/0.117/0.032 ms
4.网络模式-Host
Host
模式就是共享宿主机的网络,此时就不需要指定端口映射
-bash-4.2# docker run -d -i --name tm4 --network host centos-tomcat
3520c1b8d102bad92248d1193938aa8eabf62af456780f651c8857d7e2151992
浏览器上进行访问:
5.网络模式-None
None
模式不能与外部进行通信,只有lo
网卡:
-bash-4.2# docker run -i -d --name tm5 --network none centos-tomcat
18fe219a4f0e77f4a4cd8f756eec72cc90889f34b7def57326cceb09813c0e29
-bash-4.2# docker exec -it tm5 ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
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
三、DockerCompose
一台宿主机可能会部署多个容器,容器还可能有依赖关系,每次都手动启动容器是非常麻烦的,Compose
就是定义和运行多容器的工具,使用yml
文件配置应用程序需要的所有服务,使用docker-compose up
命令创建并启动所有服务
官网地址:https://docs.docker.com/compose
1.compose安装
直接通过yum
安装:
-bash-4.2# yum install -y docker-compose-plugin
-bash-4.2# docker compose version
Docker Compose version v2.6.0
2.官方demo入手
根据官网的demo来初次使用compose
:https://docs.docker.com/compose/gettingstarted/
2.1 创建目录
mkdir composetest
cd composetest
2.2 创建app.py
vi app.py
内容为:
import time
import redis
from flask import Flask
app = Flask(__name__)
cache = redis.Redis(host='redis', port=6379)
def get_hit_count():
retries = 5
while True:
try:
return cache.incr('hits')
except redis.exceptions.ConnectionError as exc:
if retries == 0:
raise exc
retries -= 1
time.sleep(0.5)
@app.route('/')
def hello():
count = get_hit_count()
return 'Hello World! I have been seen {} times./n'.format(count)
2.3 创建requirements.txt
vi requirements.txt
内容为:
flask
redis
2.4 创建Dockerfile
vi Dockerfile
内容为:
# syntax=docker/dockerfile:1
FROM python:3.7-alpine
WORKDIR /code
ENV FLASK_APP=app.py
ENV FLASK_RUN_HOST=0.0.0.0
RUN apk add --no-cache gcc musl-dev linux-headers
COPY requirements.txt requirements.txt
RUN pip install -r requirements.txt
EXPOSE 5000
COPY . .
CMD ["flask", "run"]
2.5 创建docker-compose.yml
vi docker-compose.yml
内容为:
version: "3.9"
services:
web:
build: .
ports:
- "8000:5000"
redis:
image: "redis:alpine"
2.6 一键启动
docker compose up
等待镜像拉取等操作完成后,出现如下输出,说明启动成功:
另开终端查看docker镜像和容器信息:
浏览器访问:
3.配置规则
由上面的demo可以知道,compose
的核心就是yml
文件,在yml中可以配置各个容器,以及容器的依赖关系,yml的可配置项有很多,官方文档查看:https://docs.docker.com/compose/compose-file/compose-file-v3/
常用的如下:
version: "3.9" #配置文件版本
services: #各个服务
容器名1:
build: . #dockerfile路径
image: centos #容器使用的镜像
expose: 8080 #对外开放端口
restart: always #自动重启
容器名2:
depends_on: # 依赖启动
- 容器名1
ports: #端口映射关系
- "8000:5000"
volumes: #数据卷
# - 宿主机绝对路径/卷标:容器目标路径
- db_data:/var/lib/mysql # 卷标方式
- /root/mysql:/var/lib/mysql2 # 绝对路径方式
environment: #环境配置
volumes: # 卷标配置
# 卷标名:[路径地址/{}自动生成目录]
db_data: {}
4.常用命令
docker compose
命令都需要在工作目录下进行,如上述的composetest
目录下
命令 | 描述 |
---|---|
docker compose up |
根据yml创建service 使用-f 指定yml文件,-d 守护模式 |
docker compose ps |
查看启动的服务 |
docker compose images |
查看镜像 |
docker compose stop/start |
停止和启动服务 |
docker compose down |
停止并删除服务,数据卷和network |
docker compose exec 服务名 命令 |
执行某个服务的命令 |
5.部署WP博客
创建目录:
mkdir my_wordpress
cd my_wordpress
docker-compose.yml文件:
version: "3.9"
services:
db:
image: mysql:5.7
volumes:
- db_data:/var/lib/mysql
restart: always
environment:
MYSQL_ROOT_PASSWORD: somewordpress
MYSQL_DATABASE: wordpress
MYSQL_USER: wordpress
MYSQL_PASSWORD: wordpress
wordpress:
depends_on:
- db
image: wordpress:latest
volumes:
- wordpress_data:/var/www/html
ports:
- "8000:80"
restart: always
environment:
WORDPRESS_DB_HOST: db:3306
WORDPRESS_DB_USER: wordpress
WORDPRESS_DB_PASSWORD: wordpress
WORDPRESS_DB_NAME: wordpress
volumes:
db_data: {}
wordpress_data: {}
一键启动:
docker compose up
浏览器访问:
6.水平扩展
水平扩展也很简单,一条命令即可:
docker compose up --scale 服务名=扩展数量 -d
运用:
-bash-4.2# docker compose ps
NAME COMMAND SERVICE STATUS PORTS
my_wordpress-db-1 "docker-entrypoint.s…" db running 33060/tcp
my_wordpress-wordpress-1 "docker-entrypoint.s…" wordpress running 0.0.0.0:8000->80/tcp, :::8000->80/tcp
-bash-4.2# docker compose up --scale db=4 -d
[+] Running 5/5
⠿ Container my_wordpress-db-4 St... 4.3s
⠿ Container my_wordpress-db-1 St... 4.7s
⠿ Container my_wordpress-db-2 St... 5.2s
⠿ Container my_wordpress-db-3 St... 5.1s
⠿ Container my_wordpress-wordpress-1 Started 4.3s
四、Harbor
Harbor是私有的镜像仓库,实际开发项目中,必然有自己打包的镜像,需要上传到仓库,以便于其他开发和运维人员拉取,首先来看下如何上传到Docker Hub
1.Docker Hub镜像上传
1.1 登录Docker Hub
如果没有账号,需要去注册:https://hub.docker.com/
docker login
输入账号密码后提示登陆成功:
WARNING! Your password will be stored unencrypted in /root/.docker/config.json.
Configure a credential helper to remove this warning. See
https://docs.docker.com/engine/reference/commandline/login/#credentials-store
Login Succeeded
1.2 docker tag修改镜像名
上传到docker hub有一个规则,即镜像名需要包含账号id(账户名),具体规则为:userId/镜像名
可以通过命令修改:docker tag 新镜像名 原镜像名
修改镜像名后,上传:
-bash-4.2# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
composetest_web latest 0f00ff5df20b 2 hours ago 185MB
centos-tomcat latest 8175d2379676 5 hours ago 643MB
centos-share 1.0 281ec8de6b48 9 hours ago 458MB
centos-vim 1.0 eb36a8966f6b 10 hours ago 458MB
centos-yum 1.0 ab2bd0073604 10 hours ago 366MB
tomcat-hello 1.0 ba5599c90061 14 hours ago 680MB
tomcat latest fb5657adc892 8 months ago 680MB
wordpress latest c3c92cc3dcb1 8 months ago 616MB
mysql 5.7 c20987f18b13 8 months ago 448MB
redis alpine 3900abf41552 9 months ago 32.4MB
centos latest 5d0da3dc9764 11 months ago 231MB
-bash-4.2# docker tag centos-tomcat aruba233/centos-tomcat
-bash-4.2# docker push aruba233/centos-tomcat
Using default tag: latest
The push refers to repository [docker.io/aruba233/centos-tomcat]
1b2f91c23757: Pushed
3cf600e944ba: Pushed
3cc84259d05d: Pushed
74ddd0ec08fa: Mounted from library/centos
latest: digest: sha256:c04d539be1ae8e21d9293ff7adeb74de9acbde9b96a509e3ce9037415edae408 size: 1167
上传成功后,就可以在docker hub上搜索到了:
2. 阿里云镜像上传
如果嫌docker hub网络太慢,可以使用阿里云的仓库:https://cr.console.aliyun.com/cn-hangzhou/instances/repositories
2.1 创建实例
2.2 创建命名空间
2.3 创建仓库
创建完成后会有提示如何使用:
2.4 上传镜像
根据提示操作即可
登录:
docker login --username=aliyun0098478676 registry.cn-hangzhou.aliyuncs.com
重命名镜像:
docker tag 8175 registry.cn-hangzhou.aliyuncs.com/aruba/centos-tomcat:latest
上传镜像:
docker push registry.cn-hangzhou.aliyuncs.com/aruba/centos-tomcat:latest
3. Harbor
上面两种都是公共仓库,想在自己服务器上搭建一个仓库也是必不可少的,Harbor就是由由VMware开源的Docker镜像仓库管理项目
3.1 Harbor安装
下载地址:https://github.com/goharbor/harbor/releases
下载后,传入linux系统,进行解压:
tar -xvf harbor-offline-installer-v2.6.0.tgz
mv harbor /usr/local
修改配置文件:
cd /usr/local/harbor
cp harbor.yml.tmpl harbor.yml
vi harbor.yml
修改如下3处,换成自己对应的IP:
3.2 配置证书
harbor内部使用nginx反向代理和https,所以要配置CA签名证书
3.2.1 生成CA证书
-bash-4.2# openssl genrsa -out ca.key 4096
-bash-4.2# openssl req -x509 -new -nodes -sha512 -days 3650 /
-subj "/C=CN/ST=Beijing/L=Beijing/O=example/OU=Personal/CN=192.168.42.4" /
-key ca.key -out ca.crt
3.2.2 生成服务器公私钥
-bash-4.2# openssl genrsa -out 192.168.42.4.key
-bash-4.2# openssl req -sha512 -new /
-subj "/C=CN/ST=Beijing/L=Beijing/O=example/OU=Personal/CN=192.168.42.4" /
-key 192.168.42.4.key -out 192.168.42.4.csr
3.2.3 生成一个x509 v3扩展文件
- 域名方式:
-bash-4.2# cat > v3.ext <<-EOF
authorityKeyIdentifier=keyid,issuer
basicConstraints=CA:FALSE
keyUsage = digitalSignature, nonRepudiation, keyEncipherment, dataEncipherment
extendedKeyUsage = serverAuth
subjectAltName = @alt_names
[alt_names]
DNS.1=harbor.od.com
DNS.2=harbor.od.com
DNS.3=harbor.od.com
EOF
- IP方式:
-bash-4.2# cat > v3.ext <<-EOF
authorityKeyIdentifier=keyid,issuer
basicConstraints=CA:FALSE
keyUsage = digitalSignature, nonRepudiation, keyEncipherment, dataEncipherment
extendedKeyUsage = serverAuth
subjectAltName = IP:192.168.42.4
EOF
3.2.4 生成CA签名证书
-bash-4.2# openssl x509 -req -sha512 -days 3650 /
-extfile v3.ext /
-CA ca.crt -CAkey ca.key -CAcreateserial /
-in 192.168.42.4.csr /
-out 192.168.42.4.crt
3.2.5 将CA签名证书和服务器私钥放入/data/cert/
目录
-bash-4.2# mkdir -p /data/cert/
-bash-4.2# cp 192.168.42.4.crt /data/cert/
-bash-4.2# cp 192.168.42.4.key /data/cert/
3.3 Docker配置证书
上面我们为harbor准备了证书,同样Docker也需要进行配置,因为Docker是要与仓库进行交互的
3.3.1 CA签名证书类型crt
转换为cert
-bash-4.2# openssl x509 -inform PEM -in 192.168.42.4.crt -out 192.168.42.4.cert
3.3.2 存放到Docker的证书目录
docker需要CA公钥,CA签名证书(cert
文件)与服务器私钥:
-bash-4.2# mkdir -p /etc/docker/certs.d/192.168.42.4/
-bash-4.2# cp 192.168.42.4.key /etc/docker/certs.d/192.168.42.4/
-bash-4.2# cp 192.168.42.4.cert /etc/docker/certs.d/192.168.42.4/
-bash-4.2# cp ca.crt /etc/docker/certs.d/192.168.42.4/
3.3.3 重启Docker
systemctl restart docker
3.4 初始化harbor
执行harbor的prepare
脚本,为nginx配置HTTPS,此过程会拉取nginx镜像:
-bash-4.2# ./prepare
执行harbor的install
脚本,初始化harbor:
-bash-4.2# ./install.sh
运行harbor
-bash-4.2# docker compose up
浏览器中访问:
yml中可以找到默认的登录账号密码:admin/Harbor12345
正常登录后:
3.5 部署harbor项目
harbor上创建一个项目:
再创建一个账号:
为该账号赋予管理员权限:
将该账户分配为刚刚创建项目的成员:
3.6 上传镜像到harbor
Docker登录到harbor:
-bash-4.2# docker login 192.168.42.4
上传镜像:
-bash-4.2# docker tag centos-tomcat 192.168.42.4/aruba/centos-tomcat:latest
-bash-4.2# docker push 192.168.42.4/aruba/centos-tomcat:latest
等待上传成功后,查看harbor中的镜像仓库:
五、Swarm
前面我们了解了Docker容器间通信,使用虚拟网络技术实现,但还有一个问题,对于多台服务器部署的docker容器间通信,该如何实现呢?
Swarm是Docker官方提供的一款集群管理工具,由若干个Docker管理机抽象成一个整体,一个Docker管理机又能集中管理它下面的Docker工作机,功能与k8s类似,但更轻量,功能也比k8s少些
Swarm和Redis集群一样遵循半数原则
,超过1/2(包括1/2)个Manager节点挂了,整个集群将不可用
1.集群搭建
1.1 克隆虚拟机
克隆三台虚拟机,分别设置HOSTNAME
,以便后续标识
修改/etc/sysconfig/network
:
-bash-4.2# vi /etc/sysconfig/network
内容为:
# Created by anaconda
NETWORKING=yes
HOSTNAME=manager
修改/etc/hosts
:
-bash-4.2# vi /etc/hosts
追加内容:
192.168.42.4 manager
另外两台也做相应的配置,只是主机名不同,修改完后重启
1.2 集群搭建
Manager
节点执行下面命令:
-bash-4.2# docker swarm init --advertise-addr 192.168.42.4
Swarm initialized: current node (7ltgt5p0vggy3w876p0k5xfrn) is now a manager.
To add a worker to this swarm, run the following command:
docker swarm join --token SWMTKN-1-1w3lb9uhu3hnsipp1iz7ogmvsatcon2zeufxysagd3sbf663fh-dqqdm5d5l7a96fe4g7qh96ffl 192.168.42.4:2377
To add a manager to this swarm, run 'docker swarm join-token manager' and follow the instructions.
其他节点执行上面的提示命令,加入到集群中:
-bash-4.2# docker swarm join --token SWMTKN-1-1w3lb9uhu3hnsipp1iz7ogmvsatcon2zeufxysagd3sbf663fh-dqqdm5d5l7a96fe4g7qh96ffl 192.168.42.4:2377
This node joined a swarm as a worker.
1.3 docker node ls查看集群节点
查看集群节点,可以看到一个manager status
为Leader
,即Manager
节点:
-bash-4.2# docker node ls
ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS ENGINE VERSION
7ltgt5p0vggy3w876p0k5xfrn * manager Ready Active Leader 20.10.17
rtavw2dgchlmhjmfth2uv0pnc node1 Ready Active 20.10.17
979ie68u7wn73w3nvlc315ldp node2 Ready Active 20.10.17
1.4 节点类型转换
Manager
节点可以将Worker
节点的类型进行升降级
命令 | 描述 |
---|---|
docker node promote 主机名 |
将节点升级为Manager 节点 |
docker node demote 主机名 |
将节点降级为Worker 节点 |
将一个节点升级为Manager
节点:
-bash-4.2# docker node promote node1
Node node1 promoted to a manager in the swarm.
-bash-4.2# docker node ls
ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS ENGINE VERSION
7ltgt5p0vggy3w876p0k5xfrn * manager Ready Active Leader 20.10.17
rtavw2dgchlmhjmfth2uv0pnc node1 Ready Active Reachable 20.10.17
979ie68u7wn73w3nvlc315ldp node2 Ready Active 20.10.17
2.集群半数协议(Raft一致性协议)
由于我们只有三个节点,所以要测试集群可用,只能把三个节点都上升为Manager
节点,都上升完后,关闭Leader
主机后,查看节点状态,发现有一台Worker
已经成为Leader
节点了:
-bash-4.2# docker node ls
ID HOSTNAME STATUS AVAILABILITY MANAGER STATUS ENGINE VERSION
7ltgt5p0vggy3w876p0k5xfrn manager Unknown Active Unreachable 20.10.17
rtavw2dgchlmhjmfth2uv0pnc * node1 Ready Active Leader 20.10.17
979ie68u7wn73w3nvlc315ldp node2 Ready Active Reachable 20.10.17
3.Swarm服务编排
上面compose
可以在一个宿主机上实现水平扩展,Swarm
则可以在Docker集群中实现水平扩展,容器
是Docker中的叫法,在Swarm
中,称为服务Service
Service相关命令如下:
命令 | 描述 |
---|---|
docker service create --name 名称 -p 端口映射 镜像名 |
创建一个service |
docker service ls |
列举出所有swarm服务 |
docker service logs 服务名 |
查看对应服务的日志输出 |
docker service inspect 服务名 |
查看服务详情 |
docker service ps 服务名 |
列举出对应服务的容器信息,信息包含运行的主机 |
docker service scale 服务名=扩展数量 |
水平扩展服务,将自动扩展到docker集群的节点中 |
docker service rm 服务名 |
删除服务 |
3.1 创建Service
-bash-4.2# docker service create --name tomcat-service tomcat
成功后查看服务:
-bash-4.2# docker service ls
ID NAME MODE REPLICAS IMAGE PORTS
ay4xeyz5qd1a tomcat-service replicated 1/1 tomcat:latest
再查看容器信息,发现创建在node2
上:
-bash-4.2# docker service ps tomcat-service
ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR PORTS
g90gcs36uzlv tomcat-service.1 tomcat:latest node2 Running Running 8 minutes ago
3.2 水平扩展
将service扩展到3个:
-bash-4.2# docker service scale tomcat-service=3
tomcat-service scaled to 3
overall progress: 3 out of 3 tasks
1/3: running
2/3: running
3/3: running
verify: Service converged
再次查看容器信息,发现三个节点上都创建了:
-bash-4.2# docker service ps tomcat-service
ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR PORTS
g90gcs36uzlv tomcat-service.1 tomcat:latest node2 Running Running 11 minutes ago
gh45bx3jwn57 tomcat-service.2 tomcat:latest manager Running Running about a minute ago
xfmeqdob533o tomcat-service.3 tomcat:latest node1 Running Running less than a second ago
3.3 自动创建与重启
如果手动停止或删除容器,或者遇到异常导致服务崩溃退出,Swarm
还是会自动创建启动容器,以拉起服务
随便找个节点将tomcat-service
容器停止,发现又会有一个新的容器被创建并启动:
-bash-4.2# docker ps | grep tomcat-service
58ac6e5387bf tomcat:latest "catalina.sh run" 3 minutes ago Up 3 minutes 8080/tcp tomcat-service.3.abmm4wcn8l7kabndc7jz3xset
-bash-4.2# docker stop 58ac6e5387bf
58ac6e5387bf
-bash-4.2# docker ps | grep tomcat-service
4a4902622d93 tomcat:latest "catalina.sh run" 6 seconds ago Up Less than a second 8080/tcp tomcat-service.3.mxt6k34ad4jltrnstf7qarbja
4.多机通信
Swarm
是通过VXLAN(Virtual eXtensible LAN)
实现的多机之间的通信,VXLAN(Virtual eXtensible LAN)
技术是当前最为主流的Overlay
标准,仅仅做了解即可
重新部署一个tomcat
服务:
-bash-4.2# docker service create -d --name mytomcat -p 8080:8080 tomcat
ph2b51mqz9tbcogsiiu7e71ny
-bash-4.2# docker service ps mytomcat
ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR PORTS
8yzqrwmtj41j mytomcat.1 tomcat:latest manager Running Running 6 seconds ago
从上面可知,目前服务只部署在一个节点上,Manager
节点的IP为:192.168.42.4
,使用其他节点IP在浏览器进行访问:
结果发现其他节点同样可以访问,这就是VXLAN
实现的,可以通过网络命令查看下:
-bash-4.2# docker network ls
NETWORK ID NAME DRIVER SCOPE
48ea34f8d77f bridge bridge local
2d68d63c8396 docker_gwbridge bridge local
9a32710598fb harbor_harbor bridge local
38f75ca6b94e host host local
f4cb49asob14 ingress overlay swarm
1c340dce736e none null local
发现ingress
的类型为overlay
:
共有 0 条评论