安装etcd集群

kuberntes 系统使用 etcd 存储所有数据，是最重要的组件之一，注意 etcd集群只能有奇数个节点(1,3,5...)，本文档使用3个节点做集群。

一、基础环境

软件包

etcd下载地址：

服务器

架构图

二、生成etcd证书与私钥

创建etcd配置文件

{
  "cn": "etcd",
  "hosts": [
    "127.0.0.1",
    "192.168.1.11",
    "192.168.1.12",
    "192.168.1.13",
    "etcd1",
    "etcd2",
    "etcd3"
  ],
  "key": {
    "algo": "rsa",
    "size": 2048
  },
  "names": [
    {
      "c": "cn",
      "st": "beijing",
      "l": "beijing",
      "o": "ctyun",
      "ou": "ops"
    }
  ]
}

hosts：定授权使用该证书的 etcd 节点 ip(应包含集群内所有节点的ip与hostname)
c：国家
st：省份
l：城市
o：公司
ou：部门

生成私钥、证书请求文件、证书

ca证书在前面的章节创建 kubernetes证书相关(cfssl)

cfssl gencert -ca=/opt/ssl/k8sca/ca.pem \
-ca-key=/opt/ssl/k8sca/ca-key.pem \
-config=/opt/ssl/k8sca/ca-config.json \
-profile=kubernetes /opt/ssl/etcd/etcd-csr.json | cfssljson -bare etcd

-ca：指定ca证书路径
-ca-key：指定cakey的路径
-config：指定ca证书签署策略配置文件

二、启动和配置etcd

创建etcd配置文件etcd.conf，此配置文件是定义一些变量，方便etcd.service文件中直接引用、方便后期维护。
配置文件的所有含义都在创建etcd.service配置文件后有详解。

mkdir -p /etc/etcd
vim /etc/etcd/etcd.conf
# [member]
etcd_name=etcd1
etcd_data_dir="/var/lib/etcd"
etcd_listen_peer_urls="https://192.168.1.11:2380" 
etcd_listen_client_urls="https://192.168.1.11:2379"
#[cluster]
etcd_initial_advertise_peer_urls="https://192.168.1.11:2380"
etcd_initial_cluster_token="etcd-cluster"
etcd_advertise_client_urls="https://192.168.1.11:2379"

创建etcd.service配置文件

etcd.service的变量都是引用etcd配置文件中

mkdir /var/lib/etcd
vim /usr/lib/systemd/system/etcd.service 
[unit]
description=etcd server
after=network.target
after=network-online.target
wants=network-online.target
documentation=https://github.com/coreos

[service]
type=notify
workingdirectory=${etcd_data_dir} 
environmentfile=/etc/etcd/etcd.conf
execstart=/usr/local/bin/etcd \
  --name=etcd1 \
  --cert-file=/opt/ssl/etcd/etcd.pem \
  --key-file=/opt/ssl/etcd/etcd-key.pem \
  --peer-cert-file=/opt/ssl/etcd/etcd.pem \
  --peer-key-file=/opt/ssl/etcd/etcd-key.pem \
  --trusted-ca-file=/opt/ssl/k8sca/ca.pem \
  --peer-trusted-ca-file=/opt/ssl/k8sca/ca.pem \
  --initial-advertise-peer-urls=${etcd_initial_advertise_peer_urls} \
  --listen-peer-urls=${etcd_listen_peer_urls} \
  --listen-client-urls=${etcd_listen_client_urls},http://127.0.0.1:2379 \
  --advertise-client-urls=${etcd_advertise_client_urls} \
  --initial-cluster-token=${etcd_initial_cluster_token} \
  --initial-cluster etcd1=https://192.168.1.11:2380,etcd2=https://192.168.1.12:2380,etcd3=https://192.168.1.13:2380 \
  --initial-cluster-state=new \
  --data-dir=/var/lib/etcd
restart=on-failure
restartsec=5
limitnofile=65536

[install]
wantedby=multi-user.target

--initial-cluster：集群当中的其他节点
--cert-file：etcd证书路径
--key-file：etcd私钥路径
--peer-cert-file：对等证书(双向证书)路径
--peer-key-file：对等证书(双向证书)私钥路径
--trusted-ca-file：作为客户端时的ca证书路径
--peer-trusted-ca-file：对等证书的ca证书路径
--initial-advertise-peer-urls：列出集群成员通信的url，用于通告集群其他成员
--listen-peer-urls：用于监听集群其他成员的url列表
--listen-client-urls：用于监听客户端通讯的url列表
--advertise-client-urls：通告客户端的url，用于列出所有客户端
--initial-cluster-token：etcd集群的初始集群令牌，服务器必须通过令牌才能加入etcd集群

启动etcd集群

集群所有节点都配置好配置文件，同时启动。

systemctl daemon-reload && systemctl enable etcd && systemctl start etcd

不使用变量的配置方法

前面分别创建了/etc/etcd/etcd.conf与/var/lib/systemd/system/etcd.service两个配置文件，且etcd.service引用了etcd.conf重定义的变量。

如若不想引用变量，可以不写etcd.conf。仅创建etcd.service如下

[root@etcd1 k8sca]# cat /usr/lib/systemd/system/etcd.service
[unit]
description=etcd server
after=network.target
after=network-online.target
wants=network-online.target
documentation=https://github.com/coreos

[service]
type=notify
workingdirectory=/var/lib/etcd/
execstart=/usr/local/bin/etcd \
  --name etcd1 \
  --cert-file=/opt/ssl/etcd/etcd.pem \
  --key-file=/opt/ssl/etcd/etcd-key.pem \
  --peer-cert-file=/opt/ssl/etcd/etcd.pem \
  --peer-key-file=/opt/ssl/etcd/etcd-key.pem \
  --trusted-ca-file=/opt/ssl/k8sca/ca.pem \
  --peer-trusted-ca-file=/opt/ssl/k8sca/ca.pem \
  --initial-advertise-peer-urls=https://192.168.1.11:2380 \
  --listen-peer-urls=https://192.168.1.11:2380 \
  --listen-client-urls=https://192.168.1.11:2379,http://127.0.0.1:2379 \
  --advertise-client-urls=https://192.168.1.11:2379 \
  --initial-cluster-token=etcd-cluster-0 \
  --initial-cluster etcd1=https://192.168.1.11:2380,etcd2=https://192.168.1.12:2380,etcd3=https://192.168.1.13:2380 \
  --initial-cluster-state=new \
  --data-dir=/var/lib/etcd
restart=on-failure
restartsec=5
limitnofile=65536

[install]
wantedby=multi-user.target

三、测试集群状态

集群中所有主机配置均一致，仅ip地址不同。

$ etcdctl --ca-file=/opt/ssl/k8sca/ca.pem --cert-file=/opt/ssl/etcd/etcd.pem --key-file=/opt/ssl/etcd/etcd-key.pem cluster-health
member aa869cb0f2e7ed31 is healthy: got healthy result from https://192.168.1.11:2379
member b08a644fd7247c5e is healthy: got healthy result from https://192.168.1.13:2379
member bb9bd2baaebf7d95 is healthy: got healthy result from https://192.168.1.12:2379

常见问题

• publish error: etcdserver: request timed out
  我部署时先使用一台进行测试，无论如何启动etcd都无法启动，提示publish error: etcdserver: request timed out。后来发现etcd.service中制定了其他etcd主机，所以当单独启动一台是连接其他主机不通，那么etcd就启动失败。所以要把etcd集群中所有主机都配置好，同时启动就ok了。

参考

http://blog.51cto.com/sgk2011/2108542
https://github.com/gjmzj/kubeasz/blob/master/docs/setup/02-install_etcd.md
https://blog.csdn.net/qq_33199919/article/details/80623055
https://skyao.gitbooks.io/learning-etcd3/content/