ITOO系统-K8s部署（二）

02-安装etcd集群.md

下载etcd/etcdctl 二进制文件、创建证书目录

创建etcd证书请求 etcd-csr.json.j2

  首先判断下是否etcd 证书已经存在，如果已经存在就跳过证书生成步骤

{
  "CN": "etcd",
  "hosts": [
    "127.0.0.1",
    "{{ inventory_hostname }}"
  ],
  "key": {
    "algo": "rsa",
    "size": 2048
  },
  "names": [
    {
      "C": "CN",
      "ST": "HangZhou",
      "L": "XS",
      "O": "k8s",
      "OU": "System"
    }
  ]
}

  etcd使用对等证书，hosts 字段必须指定授权使用该证书的 etcd 节点 IP

创建证书和私钥

cd /etc/etcd/ssl && {{ bin_dir }}/cfssl gencert \
        -ca={{ ca_dir }}/ca.pem \
        -ca-key={{ ca_dir }}/ca-key.pem \
        -config={{ ca_dir }}/ca-config.json \
        -profile=kubernetes etcd-csr.json | {{ bin_dir }}/cfssljson -bare etcd

创建etcd 服务文件 etcd.service.j2

  先创建工作目录 /var/lib/etcd/

[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={{ bin_dir }}/etcd \
  --name={{ NODE_NAME }} \
  --cert-file=/etc/etcd/ssl/etcd.pem \
  --key-file=/etc/etcd/ssl/etcd-key.pem \
  --peer-cert-file=/etc/etcd/ssl/etcd.pem \
  --peer-key-file=/etc/etcd/ssl/etcd-key.pem \
  --trusted-ca-file={{ ca_dir }}/ca.pem \
  --peer-trusted-ca-file={{ ca_dir }}/ca.pem \
  --initial-advertise-peer-urls=https://{{ inventory_hostname }}:2380 \
  --listen-peer-urls=https://{{ inventory_hostname }}:2380 \
  --listen-client-urls=https://{{ inventory_hostname }}:2379,http://127.0.0.1:2379 \
  --advertise-client-urls=https://{{ inventory_hostname }}:2379 \
  --initial-cluster-token=etcd-cluster-0 \
  --initial-cluster={{ ETCD_NODES }} \
  --initial-cluster-state=new \
  --data-dir=/var/lib/etcd
Restart=on-failure
RestartSec=5
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target

  完整参数列表请使用 etcd –help 查询

启动etcd服务

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

验证etcd集群状态

  systemctl status etcd 查看服务状态
  journalctl -u etcd 查看运行日志
  在任一 etcd 集群节点上执行如下命令

# 根据hosts中配置设置shell变量 $NODE_IPS，这里是三台
export NODE_IPS="192.168.1.1 192.168.1.2 192.168.1.3"
$ for ip in ${NODE_IPS}; do
  ETCDCTL_API=3 etcdctl \
  --endpoints=https://${ip}:2379  \
  --cacert=/etc/kubernetes/ssl/ca.pem \
  --cert=/etc/etcd/ssl/etcd.pem \
  --key=/etc/etcd/ssl/etcd-key.pem \
  endpoint health; done

  结果：

https://192.168.1.1:2379 is healthy: successfully committed proposal: took = 2.210885ms
https://192.168.1.2:2379 is healthy: successfully committed proposal: took = 2.784043ms
https://192.168.1.3:2379 is healthy: successfully committed proposal: took = 3.275709ms

03-安装docker服务.md

创建docker的systemd unit文件

[Unit]
Description=Docker Application Container Engine
Documentation=http://docs.docker.io

[Service]
Environment="PATH={{ bin_dir }}:/bin:/sbin:/usr/bin:/usr/sbin"
ExecStart={{ bin_dir }}/dockerd
ExecStartPost=/sbin/iptables -I FORWARD -s 0.0.0.0/0 -j ACCEPT
ExecReload=/bin/kill -s HUP $MAINPID
Restart=on-failure
RestartSec=5
LimitNOFILE=infinity
LimitNPROC=infinity
LimitCORE=infinity
Delegate=yes
KillMode=process

[Install]
WantedBy=multi-user.target

  dockerd 运行时会调用其它 docker 命令，如 docker-proxy，所以需要将 docker 命令所在的目录加到 PATH 环境变量中；
  docker 从 1.13 版本开始，将iptables 的filter 表的FORWARD 链的默认策略设置为DROP，从而导致 ping 其它 Node 上的 Pod IP 失败，因此必须在 filter 表的FORWARD 链增加一条默认允许规则 iptables -I FORWARD -s 0.0.0.0/0 -j ACCEPT
  运行dockerd –help 查看所有可配置参数，确保默认开启 –iptables 和 –ip-masq 选项

配置国内镜像加速

  从国内下载docker官方仓库镜像非常缓慢，所以对于k8s集群来说配置镜像加速非常重要，配置 /etc/docker/daemon.json

{
  "registry-mirrors": ["https://registry.docker-cn.com"],
  "max-concurrent-downloads": 10,
  "log-driver": "json-file",
  "log-level": "warn",
  "log-opts": {
    "max-size": "10m",
    "max-file": "3"
    }
}

  由于K8S的官方镜像存放在gcr.io仓库，因此这个镜像加速对K8S的官方镜像没有效果；好在Docker Hub上有很多K8S镜像的转存，而Docker Hub上的镜像可以加速。

清理 iptables

iptables -F && iptables -X \
        && iptables -F -t nat && iptables -X -t nat \
        && iptables -F -t raw && iptables -X -t raw \
        && iptables -F -t mangle && iptables -X -t mangle

  calico 网络支持 network-policy，使用的calico-kube-controllers 会使用到iptables 所有的四个表 filter nat raw mangle，所以一并清理。

启动 docker

安装docker查询镜像 tag的小工具

$ docker-tag library/ubuntu
"14.04"
"16.04"
"17.04"
"latest"
"trusty"
"trusty-20171117"
"xenial"
"xenial-20171114"
"zesty"
"zesty-20171114"
$ docker-tag mirrorgooglecontainers/kubernetes-dashboard-amd64
"v0.1.0"
"v1.0.0"
"v1.0.0-beta1"
"v1.0.1"
"v1.1.0-beta1"
"v1.1.0-beta2"
"v1.1.0-beta3"
"v1.7.0"
"v1.7.1"
"v1.8.0"

  需要先apt安装轻量JSON处理程序 jq
  然后下载脚本即可使用
  脚本很简单，就一行命令如下

#!/bin/bash
curl -s -S "https://registry.hub.docker.com/v2/repositories/$@/tags/" | jq '."results"[]["name"]' |sort

  对于 CentOS7 安装 jq 稍微费力一点，需要启用 EPEL 源

wget http://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
rpm -ivh epel-release-latest-7.noarch.rpm
yum install jq

验证
  运行ansible-playbook 03.docker.yml 成功后可以验证

systemctl status docker     # 服务状态
journalctl -u docker        # 运行日志
docker version
docker info

  iptables-save|grep FORWARD 查看 iptables filter表 FORWARD链，最后要有一个 -A FORWARD -j ACCEPT 保底允许规则。

iptables-save|grep FORWARD
:FORWARD ACCEPT [0:0]
:FORWARD DROP [0:0]
-A FORWARD -j DOCKER-USER
-A FORWARD -j DOCKER-ISOLATION
-A FORWARD -o docker0 -m conntrack --ctstate RELATED,ESTABLISHED -j ACCEPT
-A FORWARD -o docker0 -j DOCKER
-A FORWARD -i docker0 ! -o docker0 -j ACCEPT
-A FORWARD -i docker0 -o docker0 -j ACCEPT
-A FORWARD -j ACCEPT