截止2015年9月1日，CentOS 已经把 Kubernetes 加入官方源，所以现在安装Kubernetes已经方便很多。

各组件版本如下：

Kubernetes-1.03  docker-1.8.2  flannel-0.5.3  etcd-2.1.1

Kubernetes部署环境角色如下：

CentOS 7.2 64位系统，3台虚拟机：  master：192.168.32.15  minion1:192.168.32.16  minion2:192.168.32.17

1. 预处理

每台机器禁用iptables 避免和docker 的iptables冲突：

systemctl stop firewalld  systemctl disable firewalld

禁用selinux：

vim /etc/selinux/config  #SELINUX=enforcing  SELINUX=disabled

在2个minions机器安装docker：

yum -y install docker  yum -y update  reboot

CentOS系统，使用devicemapper作为存储后端，初始安装docker 会使用loopback, 导致docker启动报错，需要update之后再启动。

2. master结点的安装与配置

安装etcd与kubernetes-master：

yum -y install etcd kubernetes-master

修改etcd配置文件：

# egrep -v “^#” /etc/etcd/etcd.conf  ETCD_NAME=default  ETCD_DATA_DIR="/var/lib/etcd/default.etcd"  ETCD_LISTEN_CLIENT_URLS="http://0.0.0.0:2379"  ETCD_ADVERTISE_CLIENT_URLS="http://192.168.32.15:2379"

修改kube-master配置文件：

# egrep -v ‘^#’ /etc/kubernetes/apiserver | grep -v ‘^$’  KUBE_API_ADDRESS="--address=0.0.0.0"  KUBE_ETCD_SERVERS="--etcd_servers=http://192.168.32.15:2379"  KUBE_SERVICE_ADDRESSES="--service-cluster-ip-range=10.254.0.0/16"  KUBE_ADMISSION_CONTROL="--admission_control=NamespaceLifecycle,NamespaceExists,LimitRanger,SecurityContextDeny,ServiceAccount,ResourceQuota"  KUBE_API_ARGS=""

# egrep -v ‘^#’ /etc/kubernetes/controller-manager |grep -v ‘^$’  KUBE_CONTROLLER_MANAGER_ARGS="--node-monitor-grace-period=10s --pod-eviction-timeout=10s"

[root@localhost ~]# egrep -v ‘^#’ /etc/kubernetes/config | egrep -v ‘^$’  KUBE_LOGTOSTDERR="--logtostderr=true"  KUBE_LOG_LEVEL="--v=0"  KUBE_ALLOW_PRIV="--allow_privileged=false"  KUBE_MASTER="--master=http://192.168.32.15:8080"

启动服务：

systemctl enable etcd kube-apiserver kube-scheduler kube-controller-manager  systemctl start etcd kube-apiserver kube-scheduler kube-controller-manager

定义flannel网络配置到etcd,这个配置会推送到各个minions的flannel服务上：

etcdctl mk /coreos.com/network/config '{"Network":"172.17.0.0/16"}'

3. minion结点的安装与配置

yum -y install kubernetes-node flannel

修改kube-node和flannel配置文件：

# egrep -v ‘^#’ /etc/kubernetes/config | grep -v ‘^$’  KUBE_LOGTOSTDERR="--logtostderr=true"  KUBE_LOG_LEVEL="--v=0"  KUBE_ALLOW_PRIV="--allow_privileged=false"  KUBE_MASTER="--master=http://192.168.32.15:8080"

# egrep -v ‘^#’ /etc/kubernetes/kubelet | grep -v ‘^$’  KUBELET_ADDRESS="--address=127.0.0.1"  KUBELET_HOSTNAME="--hostname_override=192.168.32.16"  KUBELET_API_SERVER="--api_servers=http://192.168.32.15:8080"  KUBELET_ARGS="--pod-infra-container-image=kubernetes/pause"

为etcd服务配置flannel，修改配置文件 /etc/sysconfig/flanneld：

FLANNEL_ETCD="http://192.168.32.15:2379"  FLANNEL_ETCD_KEY="/coreos.com/network"

启动服务：

systemctl enable flanenld kubelet kube-proxy  systemctl restart flanneld docker  systemctl start kubelet kube-proxy

在每个minions可以看到2块网卡：docker0和flannel0，这2块网卡的ip在不同的机器ip地址不同：

#minion1  4: flannel0: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1472 qdisc pfifo_fast state UNKNOWN qlen 500      link/none       inet 172.17.98.0/16 scope global flannel0         valid_lft forever preferred_lft forever  5: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN       link/ether 02:42:9a:01:ca:99 brd ff:ff:ff:ff:ff:ff      inet 172.17.98.1/24 scope global docker0         valid_lft forever preferred_lft forever    #minion2  4: flannel0: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1472 qdisc pfifo_fast state UNKNOWN qlen 500      link/none       inet 172.17.67.0/16 scope global flannel0         valid_lft forever preferred_lft forever  5: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN       link/ether 02:42:25:be:ba:64 brd ff:ff:ff:ff:ff:ff      inet 172.17.67.1/24 scope global docker0         valid_lft forever preferred_lft forever

4. 检查状态

登陆master，确认minions的状态：

[root@master ~]# kubectl get nodes  NAME            LABELS                                 STATUS  192.168.32.16   kubernetes.io/hostname=192.168.32.16   Ready  192.168.32.17   kubernetes.io/hostname=192.168.32.17   Ready

kubernetes的集群就配置完成，下面就是搞pod了，后续会继续试验。