中文
Linux性能优化实战 eBPF 核心技术与实战 SDN指南 个人博客

kubeadm

Kubernetes 一键部署脚本（使用 docker 运行时）

# on master
export USE_MIRROR=true #国内用户必须使用MIRROR
git clone https://github.com/feiskyer/ops
cd ops
kubernetes/install-kubernetes.sh
# 记住控制台输出的 TOEKN 和 MASTER 地址，在其他 Node 安装时会用到

# on all nodes
git clone https://github.com/feiskyer/ops
cd ops
# Setup token and CIDR first.
# replace this with yours.
export TOKEN="xxxx"
export MASTER_IP="x.x.x.x"
export CONTAINER_CIDR="10.244.2.0/24"

# Setup and join the new node.
./kubernetes/add-node.sh

以下是详细的 kubeadm 部署集群步骤。

初始化系统

所有机器都需要初始化 docker 和 kubelet。

ubuntu

# for ubuntu 16.04
apt-get update
apt-get install -y apt-transport-https ca-certificates curl software-properties-common
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | apt-key add -
add-apt-repository "deb https://download.docker.com/linux/$(. /etc/os-release; echo "$ID") $(lsb_release -cs) stable"
apt-get update && apt-get install -y docker-ce=$(apt-cache madison docker-ce | grep 17.03 | head -1 | awk '{print $3}')

apt-get update && apt-get install -y apt-transport-https curl
curl https://mirrors.aliyun.com/kubernetes/apt/doc/apt-key.gpg | apt-key add -
    cat <<EOF >/etc/apt/sources.list.d/kubernetes.list
deb https://mirrors.aliyun.com/kubernetes/apt/ kubernetes-xenial main
EOF
apt-get update
apt-get install -y kubelet kubeadm kubectl

CentOS

yum install -y docker
systemctl enable docker && systemctl start docker

cat <<EOF > /etc/yum.repos.d/kubernetes.repo
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64/
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
EOF
setenforce 0
sed -i 's/SELINUX=enforcing/SELINUX=disabled/g' /etc/selinux/config
yum install -y kubelet kubeadm kubectl
systemctl enable kubelet && systemctl start kubelet

安装 master

# --api-advertise-addresses <ip-address>
# for flannel, setup --pod-network-cidr 10.244.0.0/16
kubeadm init --pod-network-cidr 10.244.0.0/16 --kubernetes-version latest

# enable schedule pods on the master
export KUBECONFIG=/etc/kubernetes/admin.conf
kubectl taint nodes --all node-role.kubernetes.io/master:NoSchedule-

如果需要修改 kubernetes 服务的配置选项，则需要创建一个 kubeadm 配置文件，其格式为

apiVersion: kubeadm.k8s.io/v1alpha3
kind: InitConfiguration
bootstrapTokens:
- token: "9a08jv.c0izixklcxtmnze7"
  description: "kubeadm bootstrap token"
  ttl: "24h"
- token: "783bde.3f89s0fje9f38fhf"
  description: "another bootstrap token"
  usages:
  - signing
  groups:
  - system:anonymous
nodeRegistration:
  name: "ec2-10-100-0-1"
  criSocket: "/var/run/dockershim.sock"
  taints:
  - key: "kubeadmNode"
    value: "master"
    effect: "NoSchedule"
  kubeletExtraArgs:
    cgroupDriver: "cgroupfs"
apiEndpoint:
  advertiseAddress: "10.100.0.1"
  bindPort: 6443
---
apiVersion: kubeadm.k8s.io/v1alpha3
kind: ClusterConfiguration
etcd:
  # one of local or external
  local:
    image: "k8s.gcr.io/etcd-amd64:3.2.18"
    dataDir: "/var/lib/etcd"
    extraArgs:
      listen-client-urls: "http://10.100.0.1:2379"
    serverCertSANs:
    -  "ec2-10-100-0-1.compute-1.amazonaws.com"
    peerCertSANs:
    - "10.100.0.1"
  external:
    endpoints:
    - "10.100.0.1:2379"
    - "10.100.0.2:2379"
    caFile: "/etcd/kubernetes/pki/etcd/etcd-ca.crt"
    certFile: "/etcd/kubernetes/pki/etcd/etcd.crt"
    certKey: "/etcd/kubernetes/pki/etcd/etcd.key"
networking:
  serviceSubnet: "10.96.0.0/12"
  podSubnet: "10.100.0.1/24"
  dnsDomain: "cluster.local"
kubernetesVersion: "v1.12.0"
controlPlaneEndpoint: "10.100.0.1:6443"
apiServerExtraArgs:
  authorization-mode: "Node,RBAC"
controlManagerExtraArgs:
  node-cidr-mask-size: 20
schedulerExtraArgs:
  address: "10.100.0.1"
apiServerExtraVolumes:
- name: "some-volume"
  hostPath: "/etc/some-path"
  mountPath: "/etc/some-pod-path"
  writable: true
  pathType: File
controllerManagerExtraVolumes:
- name: "some-volume"
  hostPath: "/etc/some-path"
  mountPath: "/etc/some-pod-path"
  writable: true
  pathType: File
schedulerExtraVolumes:
- name: "some-volume"
  hostPath: "/etc/some-path"
  mountPath: "/etc/some-pod-path"
  writable: true
  pathType: File
apiServerCertSANs:
- "10.100.1.1"
- "ec2-10-100-0-1.compute-1.amazonaws.com"
certificatesDir: "/etc/kubernetes/pki"
imageRepository: "k8s.gcr.io"
unifiedControlPlaneImage: "k8s.gcr.io/controlplane:v1.12.0"
auditPolicy:
  # https://kubernetes.io/docs/tasks/debug-application-cluster/audit/#audit-policy
  path: "/var/log/audit/audit.json"
  logDir: "/var/log/audit"
  logMaxAge: 7 # in days
featureGates:
  selfhosting: false
clusterName: "example-cluster"

注意：JoinConfiguration 重命名自 v1alpha2 API 中的 NodeConfiguration，而 InitConfiguration 重命名自 v1alpha2 API 中的 MasterConfiguration。

然后，在初始化 master 的时候指定 kubeadm.yaml 的路径：

kubeadm init --config ./kubeadm.yaml

配置 Network plugin

CNI bridge

mkdir -p /etc/cni/net.d
cat >/etc/cni/net.d/10-mynet.conf <<-EOF
{
    "cniVersion": "0.3.0",
    "name": "mynet",
    "type": "bridge",
    "bridge": "cni0",
    "isGateway": true,
    "ipMasq": true,
    "ipam": {
        "type": "host-local",
        "subnet": "10.244.0.0/16",
        "routes": [
            {"dst": "0.0.0.0/0"}
        ]
    }
}
EOF
cat >/etc/cni/net.d/99-loopback.conf <<-EOF
{
    "cniVersion": "0.3.0",
    "type": "loopback"
}
EOF

flannel

注意：需要 kubeadm init 时设置 --pod-network-cidr=10.244.0.0/16

kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/v0.10.0/Documentation/kube-flannel.yml

weave

sysctl net.bridge.bridge-nf-call-iptables=1
kubectl apply -f "https://cloud.weave.works/k8s/net?k8s-version=$(kubectl version | base64 | tr -d '\n')"

calico

注意：需要 kubeadm init 时设置 --pod-network-cidr=192.168.0.0/16

kubectl apply -f https://docs.projectcalico.org/v3.1/getting-started/kubernetes/installation/hosted/rbac-kdd.yaml
kubectl apply -f https://docs.projectcalico.org/v3.1/getting-started/kubernetes/installation/hosted/kubernetes-datastore/calico-networking/1.7/calico.yaml

添加 Node

kubeadm join --token <token> <master-ip>:<master-port> --discovery-token-ca-cert-hash sha256:<hash>

跟 Master 一样，添加 Node 的时候也可以自定义 Kubernetes 服务的选项，格式为

apiVersion: kubeadm.k8s.io/v1alpha2
caCertPath: /etc/kubernetes/pki/ca.crt
clusterName: kubernetes
discoveryFile: ""
discoveryTimeout: 5m0s
discoveryToken: abcdef.0123456789abcdef
discoveryTokenAPIServers:
- kube-apiserver:6443
discoveryTokenUnsafeSkipCAVerification: true
kind: NodeConfiguration
nodeRegistration:
  criSocket: /var/run/dockershim.sock
  name: thegopher
tlsBootstrapToken: abcdef.0123456789abcdef
token: abcdef.0123456789abcdef

在把 Node 加入集群的时候，指定 NodeConfiguration 配置文件的路径

kubeadm join --config ./nodeconfig.yml --token $token ${master_ip}

Cloud Provider

默认情况下，kubeadm 不包括 Cloud Provider 的配置，在 Azure 或者 AWS 等云平台上运行时，还需要配置 Cloud Provider。如

kind: MasterConfiguration
apiVersion: kubeadm.k8s.io/v1alpha2
apiServerExtraArgs:
  cloud-provider: "{cloud}"
  cloud-config: "{cloud-config-path}"
apiServerExtraVolumes:
- name: cloud
  hostPath: "{cloud-config-path}"
  mountPath: "{cloud-config-path}"
controllerManagerExtraArgs:
  cloud-provider: "{cloud}"
  cloud-config: "{cloud-config-path}"
controllerManagerExtraVolumes:
- name: cloud
  hostPath: "{cloud-config-path}"
  mountPath: "{cloud-config-path}"

删除安装

# drain and delete the node first
kubectl drain <node name> --delete-local-data --force --ignore-daemonsets
kubectl delete node <node name>

# then reset kubeadm
kubeadm reset

动态升级

kubeadm v1.8 开始支持动态升级，升级步骤为

  • 首先上传 kubeadm 配置，如 kubeadm config upload from-flags [flags]（使用命令行参数）或 kubeadm config upload from-file --config [config]（使用配置文件）

  • 在 master 上检查新版本 kubeadm upgrade plan， 当有新版本（如 v1.8.0）时，执行 kubeadm upgrade apply v1.8.0 升级控制平面

  • 手动 升级 CNI 插件（如果有新版本的话）

  • 添加自动证书回滚的 RBAC 策略 kubectl create clusterrolebinding kubeadm:node-autoapprove-certificate-rotation --clusterrole=system:certificates.k8s.io:certificatesigningrequests:selfnodeclient --group=system:nodes

  • 最后升级 kubelet

$ kubectl drain $HOST --ignore-daemonsets

# 升级软件包
$ apt-get update
$ apt-get upgrade
# CentOS 上面执行 yum 升级
# $ yum update

$ kubectl uncordon $HOST

手动升级

kubeadm v1.7 以及以前的版本不支持动态升级，但可以手动升级。

升级 Master

假设你已经有一个使用 kubeadm 部署的 Kubernetes v1.6 集群，那么升级到 v1.7 的方法为：

  1. 升级安装包 apt-get upgrade && apt-get update

  2. 重启 kubelet systemctl restart kubelet

  3. 删除 kube-proxy DaemonSet KUBECONFIG=/etc/kubernetes/admin.conf kubectl delete daemonset kube-proxy -n kube-system

  4. kubeadm init --skip-preflight-checks --kubernetes-version v1.7.1

  5. 更新 CNI 插件

升级 Node

  1. 升级安装包 apt-get upgrade && apt-get update

  2. 重启 kubelet systemctl restart kubelet

安全选项

默认情况下，kubeadm 会开启 Node 客户端证书的自动批准，如果不需要的话可以选择关闭，关闭方法为

$ kubectl delete clusterrole kubeadm:node-autoapprove-bootstrap

关闭后，增加新的 Node 时，kubeadm join 会阻塞等待管理员手动批准，匹配方法为

$ kubectl get csr
NAME                                                   AGE       REQUESTOR                 CONDITION
node-csr-c69HXe7aYcqkS1bKmH4faEnHAWxn6i2bHZ2mD04jZyQ   18s       system:bootstrap:878f07   Pending

$ kubectl certificate approve node-csr-c69HXe7aYcqkS1bKmH4faEnHAWxn6i2bHZ2mD04jZyQ
certificatesigningrequest "node-csr-c69HXe7aYcqkS1bKmH4faEnHAWxn6i2bHZ2mD04jZyQ" approved

$ kubectl get csr
NAME                                                   AGE       REQUESTOR                 CONDITION
node-csr-c69HXe7aYcqkS1bKmH4faEnHAWxn6i2bHZ2mD04jZyQ   1m        system:bootstrap:878f07   Approved,Issued

参考文档

上一页集群部署下一页kops

最后更新于