CentOS7.2中使用Kubernetes（k8s)1.4.6源碼搭建k8s容器集群環境

一、相關准備工作

1.1、准備工作

• 准備至少兩台已安裝好CentOS7.2操作系統的物理機或者虛擬機（本文配置時使用的是三台KVM虛擬機）；
• 設置hostname命令：
  hostnamectl set-hostname k8s-mst

Role	IP	Hostname
Master	192.168.3.87	k8s-mst
Node	192.168.3.88	k8s-nod1
Node	192.168.3.89	k8s-nod2

• 修改所有虛擬機或者物理機的hosts文件（/etc/hosts），添加如下內容：

192.168.3.87 k8s-mst
192.168.3.88 k8s-nod1
192.168.3.89 k8s-nod2

• 為了避免和Docker的iptables產生沖突，需要關閉Node節點上的防火牆
  systemctl stop firewalld
systemctl disable firewalld

• 為了讓各個節點的時間保持一致，需要為所有節點安裝NTP
  yum -y install ntp
systemctl start ntpd
systemctl enable ntpd

1.2、編譯Kubernetes源碼(建議在Node節點上）

1.2.1、安裝docker-engine

官方指導資料

使用以下方法可以安裝較新版本

• 添加yum庫

sudo tee /etc/yum.repos.d/docker.repo <<-'EOF'
[dockerrepo]
name=Docker Repository
baseurl=https://yum.dockerproject.org/repo/experimental/centos/7/
enabled=1
gpgcheck=1
gpgkey=https://yum.dockerproject.org/gpg
EOF

• 安裝Docker-engine
  sudo yum install -y docker-engine

• 運行Docker Daemon
  sudo systemctl start docker

• 可以使用sudo systemctl status docker查看docker Daemon的運行狀態

1.2.2、安裝Golang

sudo yum install -y golang

1.2.3、下載Kubernetes

• 從Kubernetes的github下載源碼
  git clone https://github.com/kubernetes/kubernetes.git
• 或者直接下載相應版本的release包https://codeload.github.com/kubernetes/kubernetes/tar.gz/v1.4.6
• 如果使用git clone，下載完成后進入kubernetes文件夾，使用命令git checkout v1.4.6，下載release包則解壓tar -xvf kubernetes-1.4.6.tar.gz進入kubernetes文件夾；

1.2.4、編譯Kubernetes

• 修改hosts
  由於在Kubernetes編譯過程中需要pull谷歌容器庫（gcr）中的相關鏡像，故需要修改hosts進行翻牆，hosts文件參考：https://github.com/racaljk/hosts

• 修改運行平台配置參數
  根據自己的運行平台（linux/amd64)修改hack/lib/golang.sh，把KUBE_SERVER_PLATFORMS，KUBE_CLIENT_PLATFORMS和KUBE_TEST_PLATFORMS中除linux/amd64以外的其他平台注釋掉，以此來減少編譯所用時間

• 編譯源碼
  在Kubernetes根目錄下運行命令make release-skip-tests，編譯耗時相對較長

• 編譯成功之后，可執行文件在文件夾“_output”中

二、Master配置工作

2.1、安裝ectd並修改配置文件

2.1.1、安裝必要軟件etcd

yum -y install etcd

2.1.2、 修改etcd的配置文件/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"

2.1.3、運行etcd

sudo systemctl enable etcd
sudo systemctl start etcd

2.1.4、配置etcd中的網絡

etcdctl mk /atomic.io/network/config '{"Network":"172.17.0.0/16"}'

2.2、kubernetes環境配置

2.2.1、復制命令（可執行文件）

將位於_output/release-stage/server/linux-amd64/kubernetes/server/bin/目錄下的kube-apiserver、kube-controller-manager、kube-scheduler、kubectl復制到Master節點的/usr/bin/目錄下

2.2.2、創建相應的service文件以及配置文件（shell腳本）

根據自己的的配置修改MASTER_ADDRESS和ETCD_SERVERS

#!/bin/bash
# Copyright 2016 The Kubernetes Authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

MASTER_ADDRESS=${1:-"192.168.3.87"}
ETCD_SERVERS=${2:-"http://192.168.3.87:2379"}
SERVICE_CLUSTER_IP_RANGE=${3:-"10.254.0.0/16"}
ADMISSION_CONTROL=${4:-"NamespaceLifecycle,NamespaceExists,LimitRanger,SecurityContextDeny,ResourceQuota"}

cat <<EOF >/etc/kubernetes/config
# --logtostderr=true: log to standard error instead of files
KUBE_LOGTOSTDERR="--logtostderr=true"

# --v=0: log level for V logs
KUBE_LOG_LEVEL="--v=0"

# --allow-privileged=false: If true, allow privileged containers.
KUBE_ALLOW_PRIV="--allow-privileged=false"

# How the controller-manager, scheduler, and proxy find the apiserver
KUBE_MASTER="--master=${MASTER_ADDRESS}:8080"
EOF

cat <<EOF >/etc/kubernetes/apiserver
# --insecure-bind-address=127.0.0.1: The IP address on which to serve the --insecure-port.
KUBE_API_ADDRESS="--insecure-bind-address=0.0.0.0"

# --insecure-port=8080: The port on which to serve unsecured, unauthenticated access.
KUBE_API_PORT="--insecure-port=8080"

# --kubelet-port=10250: Kubelet port
NODE_PORT="--kubelet-port=10250"

# --etcd-servers=[]: List of etcd servers to watch (http://ip:port), 
# comma separated. Mutually exclusive with -etcd-config
KUBE_ETCD_SERVERS="--etcd-servers=${ETCD_SERVERS}"

# --advertise-address=<nil>: The IP address on which to advertise 
# the apiserver to members of the cluster.
KUBE_ADVERTISE_ADDR="--advertise-address=${MASTER_ADDRESS}"

# --service-cluster-ip-range=<nil>: A CIDR notation IP range from which to assign service cluster IPs. 
# This must not overlap with any IP ranges assigned to nodes for pods.
KUBE_SERVICE_ADDRESSES="--service-cluster-ip-range=${SERVICE_CLUSTER_IP_RANGE}"

# --admission-control="AlwaysAdmit": Ordered list of plug-ins 
# to do admission control of resources into cluster. 
# Comma-delimited list of: 
#   LimitRanger, AlwaysDeny, SecurityContextDeny, NamespaceExists, 
#   NamespaceLifecycle, NamespaceAutoProvision,
#   AlwaysAdmit, ServiceAccount, ResourceQuota, DefaultStorageClass
KUBE_ADMISSION_CONTROL="--admission-control=${ADMISSION_CONTROL}"

# Add your own!
KUBE_API_ARGS=""
EOF

KUBE_APISERVER_OPTS="   \${KUBE_LOGTOSTDERR}         \\
                        \${KUBE_LOG_LEVEL}           \\
                        \${KUBE_ETCD_SERVERS}        \\
                        \${KUBE_API_ADDRESS}         \\
                        \${KUBE_API_PORT}            \\
                        \${NODE_PORT}                \\
                        \${KUBE_ADVERTISE_ADDR}      \\
                        \${KUBE_ALLOW_PRIV}          \\
                        \${KUBE_SERVICE_ADDRESSES}   \\
                        \${KUBE_ADMISSION_CONTROL}   \\
						\${KUBE_API_ARGS}"


cat <<EOF >/usr/lib/systemd/system/kube-apiserver.service
[Unit]
Description=Kubernetes API Server
Documentation=https://github.com/kubernetes/kubernetes
After=network.target
After=etcd.service

[Service]
EnvironmentFile=-/etc/kubernetes/config
EnvironmentFile=-/etc/kubernetes/apiserver
ExecStart=/usr/bin/kube-apiserver ${KUBE_APISERVER_OPTS}
Restart=on-failure
Type=notify
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target
EOF

cat <<EOF >/etc/kubernetes/controller-manager
###
# The following values are used to configure the kubernetes controller-manager

# defaults from config and apiserver should be adequate

# Add your own!
KUBE_CONTROLLER_MANAGER_ARGS=""
EOF

KUBE_CONTROLLER_MANAGER_OPTS="  \${KUBE_LOGTOSTDERR} \\
                                \${KUBE_LOG_LEVEL}   \\
                                \${KUBE_MASTER}      \\
                                \${KUBE_CONTROLLER_MANAGER_ARGS}"

cat <<EOF >/usr/lib/systemd/system/kube-controller-manager.service
[Unit]
Description=Kubernetes Controller Manager
Documentation=https://github.com/kubernetes/kubernetes

[Service]
EnvironmentFile=-/etc/kubernetes/config
EnvironmentFile=-/etc/kubernetes/controller-manager
ExecStart=/usr/bin/kube-controller-manager ${KUBE_CONTROLLER_MANAGER_OPTS}
Restart=on-failure
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target
EOF

cat <<EOF >/etc/kubernetes/scheduler
###
# kubernetes scheduler config

# Add your own!
KUBE_SCHEDULER_ARGS=""
EOF

KUBE_SCHEDULER_OPTS="   \${KUBE_LOGTOSTDERR}     \\
                        \${KUBE_LOG_LEVEL}       \\
                        \${KUBE_MASTER}          \\
                        \${KUBE_SCHEDULER_ARGS}"

cat <<EOF >/usr/lib/systemd/system/kube-scheduler.service
[Unit]
Description=Kubernetes Scheduler
Documentation=https://github.com/kubernetes/kubernetes

[Service]
EnvironmentFile=-/etc/kubernetes/config
EnvironmentFile=-/etc/kubernetes/scheduler
ExecStart=/usr/bin/kube-scheduler ${KUBE_SCHEDULER_OPTS}
Restart=on-failure
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target
EOF

systemctl daemon-reload	

2.2.3、運行相應的Kubernetes命令（shell 腳本）

for svc in kube-apiserver kube-controller-manager kube-scheduler; do 
	systemctl restart $svc
	systemctl enable $svc
	systemctl status $svc 
done

三、Node配置工作

3.1、安裝flannel並修改配置文件

3.1.1、安裝必要軟件flannel

yum -y install flannel

3.1.2、 修改flannel的配置文件/etc/sysconfig/flanneld

FLANNEL_ETCD="http://192.168.3.87:2379"
FLANNEL_ETCD_KEY="/atomic.io/network"

3.1.3、運行flannel

systemctl restart flanneld
systemctl enable flanneld
systemctl status flanneld

3.1.4、上傳網絡配置

創建一個config.json文件，內容如下：

{
"Network": "172.17.0.0/16",
"SubnetLen": 24,
"Backend": {
     "Type": "vxlan",
     "VNI": 7890
     }
 }

然后將配置上傳到etcd服務器上：
curl -L http://192.168.3.87:2379/v2/keys/atomic.io/network/config -XPUT --data-urlencode [e-mail]

3.2、kubernetes環境配置

3.2.1、復制命令（可執行文件）

將位於_output/release-stage/server/linux-amd64/kubernetes/server/bin/目錄下的kube-proxy、kubelet 復制到Node節點的/usr/bin/目錄下

3.2.2、創建相應的service文件以及配置文件（shell腳本）

根據自己的的配置修改MASTER_ADDRESS和NODE_HOSTNAME

#!/bin/bash
# Copyright 2016 The Kubernetes Authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

MASTER_ADDRESS=${1:-"192.168.3.87"}
NODE_HOSTNAME=${2:-"k8s-nod"}

cat <<EOF >/etc/kubernetes/config
# --logtostderr=true: log to standard error instead of files
KUBE_LOGTOSTDERR="--logtostderr=true"

# --v=0: log level for V logs
KUBE_LOG_LEVEL="--v=0"

# --allow-privileged=false: If true, allow privileged containers.
KUBE_ALLOW_PRIV="--allow-privileged=false"

# How the controller-manager, scheduler, and proxy find the apiserver
KUBE_MASTER="--master=${MASTER_ADDRESS}:8080"
EOF

cat <<EOF >/etc/kubernetes/proxy
###
# kubernetes proxy config

# default config should be adequate

# Add your own!
KUBE_PROXY_ARGS=""
EOF

KUBE_PROXY_OPTS="   \${KUBE_LOGTOSTDERR} \\
                    \${KUBE_LOG_LEVEL}   \\
                    \${KUBE_MASTER}    \\
                    \${KUBE_PROXY_ARGS}"

cat <<EOF >/usr/lib/systemd/system/kube-proxy.service
[Unit]
Description=Kubernetes Proxy
After=network.target

[Service]
EnvironmentFile=-/etc/kubernetes/config
EnvironmentFile=-/etc/kubernetes/kube-proxy
ExecStart=/usr/bin/kube-proxy ${KUBE_PROXY_OPTS}
Restart=on-failure
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target
EOF

cat <<EOF >/etc/kubernetes/kubelet
# --address=0.0.0.0: The IP address for the Kubelet to serve on (set to 0.0.0.0 for all interfaces)
KUBELET__ADDRESS="--address=0.0.0.0"

# --port=10250: The port for the Kubelet to serve on. Note that "kubectl logs" will not work if you set this flag.
KUBELET_PORT="--port=10250"

# --hostname-override="": If non-empty, will use this string as identification instead of the actual hostname.
KUBELET_HOSTNAME="--hostname-override=${NODE_HOSTNAME}"

# --api-servers=[]: List of Kubernetes API servers for publishing events, 
# and reading pods and services. (ip:port), comma separated.
KUBELET_API_SERVER="--api-servers=http://${MASTER_ADDRESS}:8080"

# pod infrastructure container
KUBELET_POD_INFRA_CONTAINER="--pod-infra-container-image=registry.access.redhat.com/rhel7/pod-infrastructure:latest"

# Add your own!
KUBELET_ARGS=""
EOF

KUBE_PROXY_OPTS="   \${KUBE_LOGTOSTDERR}     \\
                    \${KUBE_LOG_LEVEL}       \\
                    \${KUBELET__ADDRESS}         \\
                    \${KUBELET_PORT}            \\
                    \${KUBELET_HOSTNAME}        \\
                    \${KUBELET_API_SERVER}   \\
                    \${KUBE_ALLOW_PRIV}      \\
					\${KUBELET_POD_INFRA_CONTAINER}\\
                    \${KUBELET_ARGS}"

cat <<EOF >/usr/lib/systemd/system/kubelet.service
[Unit]
Description=Kubernetes Kubelet
After=docker.service
Requires=docker.service

[Service]
WorkingDirectory=/var/lib/kubelet
EnvironmentFile=-/etc/kubernetes/config
EnvironmentFile=-/etc/kubernetes/kubelet
ExecStart=/usr/bin/kubelet ${KUBE_PROXY_OPTS}
Restart=on-failure

[Install]
WantedBy=multi-user.target
EOF

systemctl daemon-reload

3.2.3、運行相應的Kubernetes命令（shell腳本）

for svc in docker kubelet kube-proxy; do 
	systemctl restart $svc
	systemctl enable $svc
	systemctl status $svc
done

如果出現以下情形：

● kubelet.service - Kubernetes Kubelet
   Loaded: loaded (/usr/lib/systemd/system/kubelet.service; enabled; vendor preset: disabled)
   Active: activating (auto-restart) (Result: exit-code) since Thu 2016-12-01 17:25:42 CST; 112ms ago
 Main PID: 13311 (code=exited, status=200/CHDIR)

Dec 01 17:25:42 k8s systemd[1]: Unit kubelet.service entered failed state.
Dec 01 17:25:42 k8s systemd[1]: kubelet.service failed.

是因為不存在工作目錄/var/lib/kubelet，創建相應目錄即可。

四、驗證配置以及創建dashboard

4.1、驗證環境配置

在Master節點運行命令kubectl get nodes，輸出信息如下：

[root@k8s-mst ~]# kubectl get nodes 
 NAME      STATUS    AGE
 nod1       Ready     3h 
 nod2       Ready     3h

4.2、搭建dashboard

4.2.1、創建命名空間（namespace）kube-system

創建文件kubernetes-namespace.jason，內容如下：

{
  "kind": "Namespace",
  "apiVersion": "v1",
  "metadata": {
    "name": "kube-system"
  }
}

使用命令kubectl create -f kubernetes-namespace.jason創建kube-system命名空間。

4.2.2、創建dashboard

創建kubernetes-dashboard.yaml文件

# Configuration to deploy release version of the Dashboard UI.
# Example usage: kubectl create -f <this_file>

kind: Deployment
apiVersion: extensions/v1beta1
metadata:
  labels:
    app: kubernetes-dashboard
  name: kubernetes-dashboard
  namespace: kube-system
spec:
  replicas: 1
  selector:
    matchLabels:
      app: kubernetes-dashboard
  template:
    metadata:
      labels:
        app: kubernetes-dashboard
      # Comment the following annotaion if Dashboard must not be deployed on master
      annotations:
        scheduler.alpha.kubernetes.io/tolerations: |
          [
            {
              "key": "dedicated",
              "operator": "Equal",
              "value": "master",
              "effect": "NoSchedule"
            }
          ]
    spec:
      containers:
      - name: kubernetes-dashboard
        image: gcr.io/google_containers/kubernetes-dashboard-amd64:v1.4.2
        imagePullPolicy: Always
        ports:
        - containerPort: 9090
          protocol: TCP
        args:
          # Uncomment the following line to manually specify Kubernetes API server Host
          # If not specified, Dashboard will attempt to auto discover the API server and connect
          # to it. Uncomment only if the default does not work.
          - --apiserver-host=http://192.168.3.87:8080
        livenessProbe:
          httpGet:
            path: /
            port: 9090
          initialDelaySeconds: 30
          timeoutSeconds: 30
---
kind: Service
apiVersion: v1
metadata:
  labels:
    app: kubernetes-dashboard
  name: kubernetes-dashboard
  namespace: kube-system
spec:
  type: NodePort
  ports:
  - port: 80
    targetPort: 9090
  selector:
    app: kubernetes-dashboard

注意：

1. 修改 --apiserver-host=http://192.168.3.87:8080；
2. 其中 “ image: gcr.io/google_containers/kubernetes-dashboard-amd64:v1.4.2 ”，使用的是谷歌鏡像庫，需要Node節點翻牆才會可能正常創建。

• 使用命令kubectl create -f kubernetes-dashboard.yaml創建kubernetes-dashboard Deployment和Service。

• 如無法獲取相應的鏡像創建之后，使用命令kubectl get pod --namespace="kube-system"查看會顯示如下結果：

[root@mst ~]# kubectl get pod --namespace="kube-system"
NAME                                  READY     STATUS         RESTARTS   AGE
kubernetes-dashboard-47291540-lcuox   0/1       ErrImagePull   0          1m

• 搭建自己的鏡像庫，可參考【Docker實戰】Registry & Portus搭建詳解 - 龍隱 - 博客園 http://www.cnblogs.com/xcloudbiz/articles/5497037.html

4.2.3、使用dashboard

成功運行之后：

[root@k8s-mst ~]# kubectl get pod --namespace="kube-system"
NAME                                    READY     STATUS    RESTARTS   AGE
kubernetes-dashboard-3856900779-226mr   1/1       Running   0          2m

[root@k8s-mst ~]# kubectl describe  pod kubernetes-dashboard-3856900779-226mr --namespace="kube-system"
Name:		kubernetes-dashboard-3856900779-226mr
Namespace:	kube-system
Node:		nod1/192.168.3.91
Start Time:	Tue, 22 Nov 2016 20:33:04 +0800
Labels:		app=kubernetes-dashboard
		pod-template-hash=3856900779
Status:		Running
IP:		172.18.0.2
Controllers:	ReplicaSet/kubernetes-dashboard-3856900779
Containers:
  kubernetes-dashboard:
    Container ID:	docker://d36cff522129c73f0de370857124697659662c99d370af548a1367604bac7014
    Image:		gcr.io/google_containers/kubernetes-dashboard-amd64:v1.4.2
    Image ID:		docker://sha256:c0e4ba8968ee756368cbe5f64f39b0ef8e128de90d0bdfe1d040f0773055e68a
    Port:		9090/TCP
    Args:
      --apiserver-host=http://192.168.3.87:8080
    State:			Running
      Started:			Tue, 22 Nov 2016 20:35:00 +0800
    Ready:			True
    Restart Count:		0
    Liveness:			http-get http://:9090/ delay=30s timeout=30s period=10s #success=1 #failure=3
    Volume Mounts:		<none>
    Environment Variables:	<none>
Conditions:
  Type		Status
  Initialized 	True 
  Ready 	True 
  PodScheduled 	True 
No volumes.
QoS Class:	BestEffort
Tolerations:	dedicated=master:Equal:NoSchedule
Events:
  FirstSeen	LastSeen	Count	From			SubobjectPath				Type		Reason			Message
  ---------	--------	-----	----			-------------				--------	------			-------
  4m		4m		1	{default-scheduler }						Normal		Scheduled		Successfully assigned kubernetes-dashboard-3856900779-226mr to nod1
  3m		3m		1	{kubelet nod1}		spec.containers{kubernetes-dashboard}	Normal		Pulling			pulling image "gcr.io/google_containers/kubernetes-dashboard-amd64:v1.4.2"
  3m		2m		2	{kubelet nod1}							Warning		MissingClusterDNS	kubelet does not have ClusterDNS IP configured and cannot create Pod using "ClusterFirst" policy. Falling back to DNSDefault policy.
  2m		2m		1	{kubelet nod1}		spec.containers{kubernetes-dashboard}	Normal		Pulled			Successfully pulled image "gcr.io/google_containers/kubernetes-dashboard-amd64:v1.4.2"
  2m		2m		1	{kubelet nod1}		spec.containers{kubernetes-dashboard}	Normal		Created			Created container with docker id d36cff522129; Security:[seccomp=unconfined]
  2m		2m		1	{kubelet nod1}		spec.containers{kubernetes-dashboard}	Normal		Started			Started container with docker id d36cff522129
  
[root@mst ~]# kubectl describe service  kubernetes-dashboard  --namespace="kube-system"
Name:			kubernetes-dashboard
Namespace:		kube-system
Labels:			app=kubernetes-dashboard
Selector:		app=kubernetes-dashboard
Type:			NodePort
IP:			10.254.196.154
External IPs:		192.168.3.87
Port:			<unset>	80/TCP
NodePort:		<unset>	31437/TCP
Endpoints:		172.18.0.2:9090
Session Affinity:	None

成功運行之后，就可以使用瀏覽器訪問192.168.3.87:8080/ui/使用dashboard。