一、目标

在centos7.6上安装配置etcd集群

二、架构

ETCD集群3个节点的架构
ip	hostname	os & module	Cluster Name
10.1.1.5	etcd5	centos 7.6 etcd-v3.3.2-linux-amd64	etcd-1
10.1.1.6	etcd6	centos 7.6 etcd-v3.3.2-linux-amd64	etcd-2
10.1.1.7	etcd7	centos 7.6 etcd-v3.3.2-linux-amd64	etcd-3

三、初始化环境

以下在所有节点上都要操作

1.关闭防火墙和selinux

systemctl restart ntpd
systemctl enable ntpd
systemctl stop firewalld
systemctl disable firewalld
setenforce 0

2.下载etcd软件包

wget -P /usr/local/src/ https://github.com/coreos/etcd/releases/download/v3.3.2/etcd-v3.3.2-linux-amd64.tar.gz

cd /usr/local/src
tar -zxf etcd-v3.3.2-linux-amd64.tar.gz
mv etcd-v3.3.2-linux-amd64 /usr/local/etcd-v3.3.2

3.修改系统环境变量

vim ~/.bashrc

在~/.bashrc文件的末尾追加以下两行内容

export PATH=$PATH:/usr/local/etcd-v3.3.2/
export ETCDCTL_API=3

让配置生效

source ~/.bashrc

四、配置etcd

以下需要再三个节点上操作，配置文件略有不同，注意看

1.创建etcd启动所需的配置文件

说明：配置文件基本相同，唯独advertise-client-urls和initial-advertise-peer-urls要写自己的ip地址。

● 在节点1(10.1.1.5)创建配置文件etcd.conf

vim /usr/local/etcd-v3.3.2/etcd.conf

name: etcd-1
data-dir: /usr/local/etcd-v3.3.2/data
listen-client-urls: http://0.0.0.0:2379
advertise-client-urls: http://10.1.1.5:2379
listen-peer-urls: http://0.0.0.0:2380
initial-advertise-peer-urls: http://10.1.1.5:2380
initial-cluster: etcd-1=http://10.1.1.5:2380,etcd-2=http://10.1.1.6:2380,etcd-3=http://10.1.1.7:2380
initial-cluster-token: etcd-cluster-my
initial-cluster-state: new

● 在节点2(10.1.1.6)创建配置文件etcd.conf

vim /usr/local/etcd-v3.3.2/etcd.conf

name: etcd-2
data-dir: /usr/local/etcd-v3.3.2/data
listen-client-urls: http://0.0.0.0:2379
advertise-client-urls: http://10.1.1.6:2379
listen-peer-urls: http://0.0.0.0:2380
initial-advertise-peer-urls: http://10.1.1.6:2380
initial-cluster: etcd-1=http://10.1.1.5:2380,etcd-2=http://10.1.1.6:2380,etcd-3=http://10.1.1.7:2380
initial-cluster-token: etcd-cluster-my
initial-cluster-state: new

● 在节点3(10.1.1.7)创建配置文件etcd.conf

vim /usr/local/etcd-v3.3.2/etcd.conf

name: etcd-3
data-dir: /usr/local/etcd-v3.3.2/data
listen-client-urls: http://0.0.0.0:2379
advertise-client-urls: http://10.1.1.7:2379
listen-peer-urls: http://0.0.0.0:2380
initial-advertise-peer-urls: http://10.1.1.7:2380
initial-cluster: etcd-1=http://10.1.1.5:2380,etcd-2=http://10.1.1.6:2380,etcd-3=http://10.1.1.7:2380
initial-cluster-token: etcd-cluster-my
initial-cluster-state: new

注释：

name: etcd-1 --->etcd集群节点名称（每台主机不能重复）
data-dir: /usr/local/etcd-v3.3.2/data --->存放数据的主目录
listen-client-urls: http://0.0.0.0:2379 --->监听客户端的url，0.0.0.0表示监听本机所有的ip地址
advertise-client-urls: http://10.1.1.5:2379 --->向etcd集群宣告自己的ip和端口，写本机ip
listen-peer-urls: http://0.0.0.0:2380 --->监听集群中其他成员的ip，0.0.0.0表示所有ip都能成为成员
initial-advertise-peer-urls: http://10.1.1.5:2380 --->宣告给集群中的其他成员自己的ip和端口，写本机ip
initial-cluster: etcd-1=http://10.1.1.5:2380,etcd-2=http://10.1.1.6:2380,etcd-3=http://10.1.1.7:2380 --->将集群成员的所有集群名(etcd.conf中的name值)和其ip+端口都写在这里，每个成员用逗号分隔。
initial-cluster-token: etcd-cluster-my --->集群的token，只有相同的token才能有资格成为etcd成员
initial-cluster-state: new --->集群初始化状态。new表示新建集群。当谁想加入本已存在的集群时后，那它的状态应写成existing

2.在所有节点上都创建数据存储目录

mkdir /usr/local/etcd-v3.3.2/data

3.在所有节点上都创建系统启动菜单

vim /usr/lib/systemd/system/etcd.service

[Unit]
Description=Etcd Server
After=network.target
After=network-online.target
Wants=network-online.target

[Service]
Type=notify
WorkingDirectory=/usr/local/etcd-v3.3.2
# User=etcd
ExecStart=/usr/local/etcd-v3.3.2/etcd --config-file /usr/local/etcd-v3.3.2/etcd.conf
Restart=on-failure
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target

4.首次启动etcd集群节点

启动大概原则：先启动任意一个节点，待第一个节点启动成功后，再启动剩余的两个

● 启动节点1的etcd，并设置为开机自启

systemctl daemon-reload
systemctl restart etcd && systemctl enable etcd

● 启动节点2和节点3的etcd，并设置为开机自启

systemctl daemon-reload
systemctl restart etcd && systemctl enable etcd

5.检验ETCD集群

● 在任意节点上执行，观察2379和2380端口是否启动

[root@etcd5 ~]# netstat -ntlup |grep etcd
tcp6       0      0 :::2379                 :::*                    LISTEN      6417/etcd           
tcp6       0      0 :::2380                 :::*                    LISTEN      6417/etcd

● 在任意接上上执行，查看ETCD集群成员列表

[root@etcd5 ~]# etcdctl member list
71b9fac8482ec139, started, etcd-1, http://10.1.1.5:2380, http://10.1.1.5:2379
89d4b75907acdf9a, started, etcd-3, http://10.1.1.7:2380, http://10.1.1.7:2379
b22a0934e0760a75, started, etcd-2, http://10.1.1.6:2380, http://10.1.1.6:2379

注释：查到3行，并且是已启动状态。说明我们本次的3个节点上的etcd都启动成功了，而且成功的组件成群了。

● 在任意接上上执行，查看节点健康状况

[root@etcd5 ~]# etcdctl endpoint health --endpoints=10.1.1.5:2379,10.1.1.6:2379,10.1.1.7:2379
10.1.1.5:2379 is healthy: successfully committed proposal: took = 1.79041ms
10.1.1.7:2379 is healthy: successfully committed proposal: took = 3.425099ms
10.1.1.6:2379 is healthy: successfully committed proposal: took = 3.544992ms

6.小结

无论你如何重启任意节点，集群依然监控。

无论你如何重启某个节点服务器，重启后，它依然能顺利加入到ETCD集群。

五、ETCD的基本操作命令

1.创建键值对

创建一个键name值为serena

etcdctl put name serena

2.根据键查询值

查询键为name的信息

[root@etcd5 ~]# etcdctl get name
name
serena

3.根据键的前缀模糊查询键值对信息

查询键以n开头的所有键值

[root@etcd5 ~]# etcdctl get n --prefix
name
serena
name2
kahn.xiao

4.删除指定的键

#当键存在时，成功删除返回1
[root@etcd5 ~]# etcdctl del name4
1

#当键不存在时，删除返回0
[root@etcd5 ~]# etcdctl del name666
0

5.使用URL的方式存储键值

[root@etcd5 ~]# curl http://10.1.1.5:2379/v2/keys/name5 -XPUT -d value="kahn.xiao"
{"action":"set","node":{"key":"/name5","value":"kahn.xiao","modifiedIndex":19,"createdIndex":19}}

6.使用URL的方式查询键值

[root@etcd5 ~]# curl -s http://10.1.1.5:2379/v2/keys/name5
{"action":"get","node":{"key":"/name5","value":"kahn.xiao","modifiedIndex":19,"createdIndex":19}}

7.使用URL的方式查询ETCD的版本

[root@etcd5 ~]# curl -s http://10.1.1.5:2379/version
{"etcdserver":"3.3.2","etcdcluster":"3.3.0"}

8.使用URL的方式删除键值

[root@etcd5 ~]# curl -s http://10.1.1.5:2379/v2/keys/name5 -XDELETE
{"action":"delete","node":{"key":"/name5","modifiedIndex":20,"createdIndex":19},"prevNode":{"key":"/name5","value":"kahn.xiao","modifiedIndex":19,"createdIndex":19}}

9.使用URL的方式创建目录

创建一个叫kahnDIR的目录

[root@etcd5 ~]# curl -s http://10.1.1.5:2379/v2/keys/kahnDIR -XPUT -d dir=true
{"action":"set","node":{"key":"/kahnDIR","dir":true,"modifiedIndex":21,"createdIndex":21}}

10.使用URL的方式查询所有目录

[root@etcd5 ~]# curl -s http://10.1.1.5:2379/v2/keys/?recursive=true
{"action":"get","node":{"dir":true,"nodes":[{"key":"/kahnDIR","dir":true,"modifiedIndex":21,"createdIndex":21},
{"key":"/kahnDIR2","dir":true,"modifiedIndex":22,"createdIndex":22},
{"key":"/kahnDIR3","dir":true,"modifiedIndex":23,"createdIndex":23}]}}

注释：这里查到有3个目录，目录名分别叫kahnDIR,kahnDIR2,kahnDIR3

11.使用URL的方式删除一个目录

删除一个叫kahnDIR3的目录

[root@etcd5 ~]# curl -s 'http://10.1.1.5:2379/v2/keys/kahnDIR3?dir=true' -XDELETE
{"action":"delete","node":{"key":"/kahnDIR3","dir":true,"modifiedIndex":24,"createdIndex":23},"prevNode":{"key":"/kahnDIR3","dir":true,"modifiedIndex":23,"createdIndex":23}}

-------------END-------------------2020年11月3日22:59:06-------------------------

写帖子非常耗时间，非常累，如果你觉得本帖对你起到帮助作用了，那么希望能得到你的打赏鼓励，以鼓励我继续分享更多干货给你

ETCD-01_etcd集群安装配置