kubernetes集群安装指南:etcd集群部署

发布时间:2020-08-06 03:13:31 作者:清白之年
来源:网络 阅读:573

etcd 是基于 Raft 的分布式 key-value 存储系统,由 CoreOS 开发,常用于服务发现、共享配置以及并发控制(如 leader 选举、分布式锁等)。kubernetes 使用 etcd 存储所有运行数据。

部署准备

所有操作都是在devops机器上操作,将相关脚本编辑好,然后通过ansible 脚本模块统一执行。

定义环境变量
#################### Variable parameter setting ######################
ETCD_INSTALL_PATH="/data/apps/k8s/etcd"
ETCD_BIN_DIR="${ETCD_INSTALL_PATH}/bin"
ETCD_DATA_DIR="${ETCD_INSTALL_PATH}/data"
ETCD_WAL_DIR="${ETCD_INSTALL_PATH}/wal"
N01="master-k8s-n01=https://10.10.10.22:2380"
N02="master-k8s-n02=https://10.10.10.23:2380"
N03="master-k8s-n03=https://10.10.10.24:2380"
ETCD_ENPOIDTS="$N01,$N02,$N03"
CA_PATH="/etc/k8s/ssl"
SOFTWARE="/root/software"
VERSION=v3.3.13
PACKAGE="etcd-${VERSION}-linux-amd64.tar.gz"
HOSTNAME=`hostname`
DOWNLOAD_URL="https://github.com/devops-apps/download/raw/master/etcd/${PACKAG}"
ETH_INTERFACE=eth2
LISTEN_IP=$(ifconfig | grep -A 1 ${ETH_INTERFACE} |grep inet |awk '{print $2}')
USER=k8s
下载和分发 etcd 二进制文件

登录devops访问etcd 的 release 页面下载最新版本的发布包并通过ansible同步到三个master三个节点/root/software目录下:

wget $DOWNLOAD_URL -P $SOFTWARE

2 部署etcd集群

在devops主机上使用ansible分别对三个master节点执行以下操作:

2.1 安装etcd二进制执行命令
### 1.Check if the install directory exists.
if [ ! -d $ETCD_BIN_DIR ]; then
     mkdir -p $ETCD_BIN_DIR
     mkdir -p $ETCD_DATA_DIR
     mkdir -p $ETCD_WAL_DIR
fi

### 2.Install etcd binary of kubernetes.
if [ ! -f "$SOFTWARE/etcd-${VERSION}-linux-amd64.tar.gz" ]; then
     sudo wget $DOWNLOAD_URL -P $SOFTWARE >>/tmp/install.log  2>&1
fi

cd $SOFTWARE && tar -xzf etcd-${VERSION}-linux-amd64.tar.gz -C ./
cp -fp etcd-${VERSION}-linux-amd64/etcd* $ETCD_BIN_DIR
ln -sf  $ETCD_BIN_DIR/{etcd,etcdctl} /usr/local/bin
sudo chmod -R 755 $ETCD_INSTALL_PATH
sudo chown -R $USER:$USER $ETCD_INSTALL_PATH
2.2 创建 etcd 的 systemd unit 启动服务文件
cat >/usr/lib/systemd/system/etcd.service<<EOF
[Unit]
Description=Etcd Server
After=network.target
After=network-online.target
Wants=network-online.target
[Service]
Type=notify
User=${USER}
WorkingDirectory=${ETCD_DATA_DIR}
# set GOMAXPROCS to number of processors
ExecStart=/bin/bash -c "GOMAXPROCS=$(nproc) ${ETCD_BIN_DIR}/etcd  \\
                  --name=${HOSTNAME} \\
                  --data-dir=${ETCD_DATA_DIR} \\
                  --wal-dir=${ETCD_WAL_DIR} \\
                  --cert-file=${CA_PATH}/etcd.pem \\
                  --key-file=${CA_PATH}/etcd-key.pem \\
                  --trusted-ca-file=${CA_PATH}/ca.pem \\
                  --peer-cert-file=${CA_PATH}/etcd.pem \\
                  --peer-key-file=${CA_PATH}/etcd-key.pem \\
                  --peer-trusted-ca-file=${CA_PATH}/ca.pem \\
                  --peer-client-cert-auth=true \\
                  --client-cert-auth=true \\
                  --listen-peer-urls=https://${LISTEN_IP}:2380 \\
                  --initial-advertise-peer-urls=https://${LISTEN_IP}:2380 \\
                  --listen-client-urls=https://${LISTEN_IP}:2379,https://127.0.0.1:2379 \\
                  --advertise-client-urls=https://${LISTEN_IP}:2379 \\
                  --initial-cluster-token=etcd-cluster-0 \\
                  --initial-cluster=${ETCD_ENPOIDTS} \\
                  --initial-cluster-state=new \\
                  --auto-tls=true \\
                  --auto-compaction-mode=periodic \\
                  --auto-compaction-retention=1 \\
                  --max-request-bytes=33554432 \\
                  --quota-backend-bytes=6442450944 \\
                  --heartbeat-interval=250 \\
                  --election-timeout=2000 \\
                  --peer-auto-tls=true"
Restart=on-failure
RestartSec=5
LimitNOFILE=65536
[Install]
WantedBy=multi-user.target
EOF
2.3 启动 etcd 服务并检查启动状态
启动etcd服务
systemctl  etcd.service start
检查etcd启动状态
systemctl status etcd|grep Active

确保状态为 active (running),否则查看日志,确认原因:

sudo journalctl -u etcd
验证etcd服务状态及查看当前leader

验证etcd服务状态

ETCDCTL_API=3 etcdctl \
  --endpoints=${ETCD_ENPOIDTS} \
  --cacert=/etc/k8s/ssl/ca.pem \
  --cert=/etc/k8s/ssl/etcd.pem \
  --key=/etc/k8s/ssl/etcd-key.pem endpoint health

当输出值为以下状态即为正常:
https://10.10.10.22:2379 is healthy: successfully committed proposal: took = 3.693346ms
https://10.10.10.23:2379 is healthy: successfully committed proposal: took = 2.80687ms
https://10.10.10.24:2379 is healthy: successfully committed proposal: took = 4.493037ms

查看当前leader

ETCDCTL_API=3 etcdctl \
  --endpoints=${ETCD_ENPOIDTS} \
  --cacert=/etc/k8s/ssl/ca.pem \
  --cert=/etc/k8s/ssl/etcd.pem \
  --key=/etc/k8s/ssl/etcd-key.pem endpoint status

输出:
https://10.10.10.22:2379, 641670f7677ceff, 3.3.13, 2.9 MB, false, 2, 6198986
https://10.10.10.23:2379, ce37122364c277fa, 3.3.13, 2.9 MB, true, 2, 6198986
https://10.10.10.24:2379, 4d1003897626ccde, 3.3.13, 2.9 MB, false, 2, 6198986

到这里,整个etcd集群就部署完成,关于更多的etcd操作可以阅读etcd官方文档,部署好etcd数据库,开部署kubernetes master节点相关组件,请参考:kubernetes集群安装指南:apiserver节点组件部署,关于etcd脚本请从此处获取;

推荐阅读:
  1. Kubernetes集群部署的方式有哪些?
  2. kubernetes集群部署Flannel组件

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。

kubernetes etcd etcd集群部署

上一篇:Go语言中的流程控制

下一篇:Hadoop 2.4安装与配置

相关阅读

您好,登录后才能下订单哦!

密码登录
登录注册
其他方式登录
点击 登录注册 即表示同意《亿速云用户服务条款》