4.3 安装etcd

在master上安装etcd,本文对应的是192.168.100.20这台机器

成都服务器托管，创新互联公司提供包括服务器租用、成都棕树机房、带宽租用、云主机、机柜租用、主机租用托管、CDN网站加速、域名与空间等业务的一体化完整服务。电话咨询：13518219792

下载etcd安装包

wget https://github.com/coreos/etcd/releases/download/v3.2.7/etcd-v3.2.7-linux-amd64.tar.gz

解包，将etcd-v3.2.7-linux-amd64下的etcd,etcdctl到/usr/local/bin目录下

cp etcd-v3.2.7-linux-amd64/{etcd,etcdctl} /usr/local/bin

制作etcd启动脚本

cat etcd.sh

#!/bin/sh

ETCD_ADDR=http://0.0.0.0:2379

ETCD_DATA_DIR=/var/lib/etcd

ETCD_NAME=kubernetes

! test -d $ETCD_DATA_DIR && mkdir -p $ETCD_DATA_DIR

cat ? << EOF >/usr/lib/systemd/system/etcd.service

[Unit]

Description=Etcd Server

[Service]

ExecStart=/usr/local/bin/etcd --listen-client-urls='$ETCD_ADDR' --advertise-client-urls='$ETCD_ADDR'

[Install]

WantedBy=multi-user.target

EOF

systemctl daemon-reload

systemctl enable etcd

systemctl start etcd

启动etcd

systemctl start etcd.service

4.4 安装apiserver,controller-manager,scheduler,kubectl

在master上安装apiserver

apiserver,controller-manager,sheduler等组件都在kubernetes包中，所以下载kubernetes包即可

wget https://storage.googleapis.com/kubernetes-release/release/v1.6.10/kubernetes-server-linux-amd64.tar.gz

解包，并将kubernetes/server/bin下的kube-apiserver,kube-controller-manager,kube-scheduler,kubectl,kubefed 复制到/usr/local/bin下

cp kubernetes/server/bin/{kube-apiserver,kube-controller-manager,kube-scheduler,kubectl,kubefed} /usr/local/bin

制作apiserver ,controller-manager,scheduler的启动脚本

apiserver 启动脚本

cat apiserver.sh

#!/bin/sh

KUBE_LOGTOSTDERR=true

KUBE_LOG_LEVEL=2

KUBE_ETCD_SERVERS=http://192.168.100.20:2379

KUBE_CLUSTER_IP=10.10.10.0/24

INSECURE_BIND_ADDR=0.0.0.0

INSECURE_PORT=8080

LOG_DIR=/var/log/kubernetes

cat

[Unit]

Description=Kubernetes API Server

After=network.target

After=etcd.service

[Service]

ExecStart=/usr/local/bin/kube-apiserver  \\

--logtostderr=${KUBE_LOGTOSTDERR} \\

--v=${KUBE_LOG_LEVEL} \\

--log-dir='${LOG_DIR}' \\

--etcd_servers=${KUBE_ETCD_SERVERS} \\

--service-cluster-ip-range=${KUBE_CLUSTER_IP} \\

--service-node-port-range=1-65535 \\

--admission_control=NamespaceLifecycle,LimitRanger,SecurityContextDeny,ResourceQuota \\

--insecure-bind-address=${INSECURE_BIND_ADDR} \\

--insecure-port=${INSECURE_PORT}

Restart=on-failure

[Install]

WantedBy=multi-user.target

EOF

systemctl daemon-reload

systemctl enable apiserver

systemctl start apiserver

启动服务

systemctl start apiserver.service

controller-manager启动脚本

cat controller-mananger.sh

#!/bin/sh

KUBE_LOGTOSTDERR=true

KUBE_LOG_LEVEL=4kubernetes-server-linux-amd64.tar.gz

KUBE_MASTER=192.168.100.20:8080

cat

[Unit]

Description=Kubernetes Controller Manager

Documentation=https://github.com/GoogleCloudPlatform/kubernetes

After=network.target

After=apiserver.service

[Service]

ExecStart=/usr/local/bin/kube-controller-manager \\

--logtostderr=${KUBE_LOGTOSTDERR} \\

--v=${KUBE_LOG_LEVEL} \\

--log-dir="/var/log/controller" \\

--master=${KUBE_MASTER} \\

--service-cluster-ip-range=10.10.10.0/24

Restart=on-failure

[Install]

WantedBy=multi-user.target

EOF

systemctl daemon-reload

systemctl enable controller-manager

systemctl start controller-manager

启动服务

systemctl start controller-manager.service

scheduler启动脚本

cat scheduler.sh

#!bin/sh

KUBE_LOGTOSTDERR=true

KUBE_LOG_LEVEL=4

KUBE_MASTER=192.168.100.20:8080

cat ?

[Unit]

Description=Kubernetes Scheduler

Documentation=https://github.com/GoogleCloudPlatform/kubernetes

[Service]

ExecStart=/usr/local/bin/kube-scheduler \\

--logtostderr=${KUBE_LOGTOSTDERR} \\

--v=${KUBE_LOG_LEVEL} \\

--master=${KUBE_MASTER} \\

--log-dir='/var/log/scheduler'

Restart=on-failure

[Install]

WantedBy=multi-user.target

EOF

systemctl daemon-reload

systemctl enable scheduler

systemctl start scheduler

启动服务systemctl start scheduler.service

4.5 安装kubelet,kube-proxy

这两个组件安装在minion端，将在master下载的kubernetes-server-linux-amd64.tar.gz解压后的kubernetes/server/bin下的kubelet,kube-proxy复制到minion的/usr/local/bin下

cp kubernetes/server/bin/{kubelet,kube-proxy} /usr/local/bin

制作启动脚本

kubelet启动脚本

cat kubelet.sh

#!/bin/sh

KUBE_LOGTOSTDERR=true

KUBE_LOG_LEVEL=4

KUBE_SERVERS=http://192.168.100.20:8080

MINION_HOSTNAME=192.168.100.93

cat ?

[Unit]

Description=Kubernetes Kubelet

After=docker.socket cadvisor.service

Requires=docker.socket

[Service]

ExecStart=/usr/local/bin/kubelet \\

--logtostderr=${KUBE_LOGTOSTDERR} \\

--v=${KUBE_LOG_LEVEL} \\

--api_servers=${KUBE_SERVERS} \\

--hostname_override=${MINION_HOSTNAME} \\

--cluster-DNS=10.10.10.10 \\

--cluster-domain=cluster.local \\

--allow-privileged=true \\

--pod-infra-container-image=registry:5000/pause-amd64:3.0

Restart=on-failure

[Install]

WantedBy=multi-user.target

EOF

systemctl daemon-reload

systemctl enable kubelet

systemctl start kubelet

启动服务

systemctl start kubelet.service

kube-proxy启动脚本

cat proxy.sh

#!/bin/sh

KUBE_LOGTOSTDERR=true

KUBE_LOG_LEVEL=4

KUBE_SERVERS=http://192.168.100.20:8080

cat ?

[Unit]

Description=Kubernetes Proxy

# the proxy crashes if etcd isn't reachable.

# https://github.com/GoogleCloudPlatform/kubernetes/issues/1206

After=network.target

[Service]

ExecStart=/usr/local/bin/kube-proxy \\

--logtostderr=${KUBE_LOGTOSTDERR} \\

--v=${KUBE_LOG_LEVEL} \\

--master=${KUBE_SERVERS}

Restart=on-failure

[Install]

WantedBy=multi-user.target

EOF

systemctl daemon-reload

systemctl enable proxy

systemctl start proxy

启动服务systemctl start proxy.service