一 获取Kubernetes二进制文件
1.1 master节点服务
kubernetes master 节点运行如下组件:
- kube-apiserver
- kube-scheduler
- kube-controller-manager
- kube-nginx
kube-apiserver、kube-scheduler 和 kube-controller-manager 均以多实例模式运行:
kube-scheduler 和 kube-controller-manager 会自动选举产生一个 leader 实例,其它实例处于阻塞模式,当 leader 挂了后,重新选举产生新的 leader,从而保证服务可用性。
kube-apiserver 是无状态的,需要通过 kube-nginx 进行代理访问,从而保证服务可用性。
1.2 安装Kubernetes
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# wget https://storage.googleapis.com/kubernetes-release/release/v1.18.3/kubernetes-server-linux-amd64.tar.gz 3 [root@master01 work]# tar -xzvf kubernetes-server-linux-amd64.tar.gz 4 [root@master01 work]# cd kubernetes 5 [root@master01 kubernetes]# tar -xzvf kubernetes-src.tar.gz
提示:本步骤操作仅需要在master01节点操作。
1.3 分发Kubernetes
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# for master_ip in ${MASTER_IPS[@]} 4 do 5 echo ">>> ${master_ip}" 6 scp -rp kubernetes/server/bin/{apiextensions-apiserver,kube-apiserver,kube-controller-manager,kube-proxy,kube-scheduler,kubeadm,kubectl,kubelet,mounter} root@${master_ip}:/opt/k8s/bin/ 7 ssh root@${master_ip} "chmod +x /opt/k8s/bin/*" 8 done
提示:本步骤操作仅需要在master01节点操作。
二 创建kube-apiserver证书
2.1 高可用apiserver介绍
本实验部署一个三实例 kube-apiserver 集群的步骤,它们通过 kube-nginx 进行代理访问,从而保证服务可用性。
2.2 创建Kubernetes证书和私钥
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# cat > kubernetes-csr.json <<EOF 4 { 5 "CN": "kubernetes", 6 "hosts": [ 7 "127.0.0.1", 8 "172.24.8.71", 9 "172.24.8.72", 10 "172.24.8.73", 11 "172.24.8.100", 12 "${CLUSTER_KUBERNETES_SVC_IP}", 13 "kubernetes", 14 "kubernetes.default", 15 "kubernetes.default.svc", 16 "kubernetes.default.svc.cluster", 17 "kubernetes.default.svc.cluster.local." 18 ], 19 "key": { 20 "algo": "rsa", 21 "size": 2048 22 }, 23 "names": [ 24 { 25 "C": "CN", 26 "ST": "Shanghai", 27 "L": "Shanghai", 28 "O": "k8s", 29 "OU": "System" 30 } 31 ] 32 } 33 EOF #创建Kubernetes的CA证书请求文件
提示:本步骤操作仅需要在master01节点操作。
解释:
hosts 字段指定授权使用该证书的 IP 和域名列表,这里列出了 master 节点 IP、kubernetes 服务的 IP 和域名;
kubernetes 服务 IP 是 apiserver 自动创建的,一般是 --service-cluster-ip-range 参数指定的网段的第一个IP,后续可以通过下面命令获取:
# kubectl get svc kubernetes
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# cfssl gencert -ca=/opt/k8s/work/ca.pem 3 -ca-key=/opt/k8s/work/ca-key.pem -config=/opt/k8s/work/ca-config.json 4 -profile=kubernetes kubernetes-csr.json | cfssljson -bare kubernetes #生成密钥和证书
提示:本步骤操作仅需要在master01节点操作。
2.3 分发证书和私钥
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# for master_ip in ${MASTER_IPS[@]} 4 do 5 echo ">>> ${master_ip}" 6 ssh root@${master_ip} "mkdir -p /etc/kubernetes/cert" 7 scp kubernetes*.pem root@${master_ip}:/etc/kubernetes/cert/ 8 done
提示:本步骤操作仅需要在master01节点操作。
三 配置kube-apiserver审计
3.1 创建加密配置文件
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# cat > encryption-config.yaml <<EOF 4 kind: EncryptionConfig 5 apiVersion: v1 6 resources: 7 - resources: 8 - secrets 9 providers: 10 - aescbc: 11 keys: 12 - name: key1 13 secret: ${ENCRYPTION_KEY} 14 - identity: {} 15 EOF
提示:本步骤操作仅需要在master01节点操作。
3.2 分发加密配置文件
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# for master_ip in ${MASTER_IPS[@]} 4 do 5 echo ">>> ${master_ip}" 6 scp encryption-config.yaml root@${master_ip}:/etc/kubernetes/ 7 done
提示:本步骤操作仅需要在master01节点操作。
3.3 创建审计策略文件
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# cat > audit-policy.yaml <<EOF 4 apiVersion: audit.k8s.io/v1beta1 5 kind: Policy 6 rules: 7 # The following requests were manually identified as high-volume and low-risk, so drop them. 8 - level: None 9 resources: 10 - group: "" 11 resources: 12 - endpoints 13 - services 14 - services/status 15 users: 16 - 'system:kube-proxy' 17 verbs: 18 - watch 19 20 - level: None 21 resources: 22 - group: "" 23 resources: 24 - nodes 25 - nodes/status 26 userGroups: 27 - 'system:nodes' 28 verbs: 29 - get 30 31 - level: None 32 namespaces: 33 - kube-system 34 resources: 35 - group: "" 36 resources: 37 - endpoints 38 users: 39 - 'system:kube-controller-manager' 40 - 'system:kube-scheduler' 41 - 'system:serviceaccount:kube-system:endpoint-controller' 42 verbs: 43 - get 44 - update 45 46 - level: None 47 resources: 48 - group: "" 49 resources: 50 - namespaces 51 - namespaces/status 52 - namespaces/finalize 53 users: 54 - 'system:apiserver' 55 verbs: 56 - get 57 58 # Don't log HPA fetching metrics. 59 - level: None 60 resources: 61 - group: metrics.k8s.io 62 users: 63 - 'system:kube-controller-manager' 64 verbs: 65 - get 66 - list 67 68 # Don't log these read-only URLs. 69 - level: None 70 nonResourceURLs: 71 - '/healthz*' 72 - /version 73 - '/swagger*' 74 75 # Don't log events requests. 76 - level: None 77 resources: 78 - group: "" 79 resources: 80 - events 81 82 # node and pod status calls from nodes are high-volume and can be large, don't log responses for expected updates from nodes 83 - level: Request 84 omitStages: 85 - RequestReceived 86 resources: 87 - group: "" 88 resources: 89 - nodes/status 90 - pods/status 91 users: 92 - kubelet 93 - 'system:node-problem-detector' 94 - 'system:serviceaccount:kube-system:node-problem-detector' 95 verbs: 96 - update 97 - patch 98 99 - level: Request 100 omitStages: 101 - RequestReceived 102 resources: 103 - group: "" 104 resources: 105 - nodes/status 106 - pods/status 107 userGroups: 108 - 'system:nodes' 109 verbs: 110 - update 111 - patch 112 113 # deletecollection calls can be large, don't log responses for expected namespace deletions 114 - level: Request 115 omitStages: 116 - RequestReceived 117 users: 118 - 'system:serviceaccount:kube-system:namespace-controller' 119 verbs: 120 - deletecollection 121 122 # Secrets, ConfigMaps, and TokenReviews can contain sensitive & binary data, 123 # so only log at the Metadata level. 124 - level: Metadata 125 omitStages: 126 - RequestReceived 127 resources: 128 - group: "" 129 resources: 130 - secrets 131 - configmaps 132 - group: authentication.k8s.io 133 resources: 134 - tokenreviews 135 # Get repsonses can be large; skip them. 136 - level: Request 137 omitStages: 138 - RequestReceived 139 resources: 140 - group: "" 141 - group: admissionregistration.k8s.io 142 - group: apiextensions.k8s.io 143 - group: apiregistration.k8s.io 144 - group: apps 145 - group: authentication.k8s.io 146 - group: authorization.k8s.io 147 - group: autoscaling 148 - group: batch 149 - group: certificates.k8s.io 150 - group: extensions 151 - group: metrics.k8s.io 152 - group: networking.k8s.io 153 - group: policy 154 - group: rbac.authorization.k8s.io 155 - group: scheduling.k8s.io 156 - group: settings.k8s.io 157 - group: storage.k8s.io 158 verbs: 159 - get 160 - list 161 - watch 162 163 # Default level for known APIs 164 - level: RequestResponse 165 omitStages: 166 - RequestReceived 167 resources: 168 - group: "" 169 - group: admissionregistration.k8s.io 170 - group: apiextensions.k8s.io 171 - group: apiregistration.k8s.io 172 - group: apps 173 - group: authentication.k8s.io 174 - group: authorization.k8s.io 175 - group: autoscaling 176 - group: batch 177 - group: certificates.k8s.io 178 - group: extensions 179 - group: metrics.k8s.io 180 - group: networking.k8s.io 181 - group: policy 182 - group: rbac.authorization.k8s.io 183 - group: scheduling.k8s.io 184 - group: settings.k8s.io 185 - group: storage.k8s.io 186 187 # Default level for all other requests. 188 - level: Metadata 189 omitStages: 190 - RequestReceived 191 EOF
提示:本步骤操作仅需要在master01节点操作。
3.4 分发策略文件
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# for master_ip in ${MASTER_IPS[@]} 4 do 5 echo ">>> ${master_ip}" 6 scp audit-policy.yaml root@${master_ip}:/etc/kubernetes/audit-policy.yaml 7 done
提示:有关审计知识参考:https://www.cnblogs.com/liabio/p/11696081.html。
提示:本步骤操作仅需要在master01节点操作。
四 配置metrics-server
4.1 创建访问 metrics-server的证书和密钥
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# cat > proxy-client-csr.json <<EOF 3 { 4 "CN": "system:metrics-server", 5 "hosts": [], 6 "key": { 7 "algo": "rsa", 8 "size": 2048 9 }, 10 "names": [ 11 { 12 "C": "CN", 13 "ST": "Shanghai", 14 "L": "Shanghai", 15 "O": "k8s", 16 "OU": "System" 17 } 18 ] 19 } 20 EOF #创建metrics-server的CA证书请求文件
提示:本步骤操作仅需要在master01节点操作。
解释:
CN 名称需要位于 kube-apiserver 的 --requestheader-allowed-names 参数中,否则后续访问 metrics 时会提示权限不足。
1 [root@master01 work]# cfssl gencert -ca=/opt/k8s/work/ca.pem 2 -ca-key=/opt/k8s/work/ca-key.pem -config=/opt/k8s/work/ca-config.json 3 -profile=kubernetes proxy-client-csr.json | cfssljson -bare proxy-client #生成密钥和证书
提示:本步骤操作仅需要在master01节点操作。
4.2 分发证书和私钥
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# for master_ip in ${MASTER_IPS[@]} 4 do 5 echo ">>> ${master_ip}" 6 scp proxy-client*.pem root@${master_ip}:/etc/kubernetes/cert/ 7 done
提示:本步骤操作仅需要在master01节点操作。
五 配置kube-apiserver
5.1 创建kube-apiserver的systemd
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# cat > kube-apiserver.service.template <<EOF 4 [Unit] 5 Description=Kubernetes API Server 6 Documentation=https://github.com/GoogleCloudPlatform/kubernetes 7 After=network.target 8 9 [Service] 10 WorkingDirectory=${K8S_DIR}/kube-apiserver 11 ExecStart=/opt/k8s/bin/kube-apiserver \ 12 --insecure-port=0 \ 13 --secure-port=6443 \ 14 --bind-address=##MASTER_IP## \ 15 --advertise-address=##MASTER_IP## \ 16 --default-not-ready-toleration-seconds=360 \ 17 --default-unreachable-toleration-seconds=360 \ 18 --feature-gates=DynamicAuditing=true \ 19 --max-mutating-requests-inflight=2000 \ 20 --max-requests-inflight=4000 \ 21 --default-watch-cache-size=200 \ 22 --delete-collection-workers=2 \ 23 --encryption-provider-config=/etc/kubernetes/encryption-config.yaml \ 24 --etcd-cafile=/etc/kubernetes/cert/ca.pem \ 25 --etcd-certfile=/etc/kubernetes/cert/kubernetes.pem \ 26 --etcd-keyfile=/etc/kubernetes/cert/kubernetes-key.pem \ 27 --etcd-servers=${ETCD_ENDPOINTS} \ 28 --tls-cert-file=/etc/kubernetes/cert/kubernetes.pem \ 29 --tls-private-key-file=/etc/kubernetes/cert/kubernetes-key.pem \ 30 --audit-dynamic-configuration \ 31 --audit-log-maxage=30 \ 32 --audit-log-maxbackup=3 \ 33 --audit-log-maxsize=100 \ 34 --audit-log-truncate-enabled=true \ 35 --audit-log-path=${K8S_DIR}/kube-apiserver/audit.log \ 36 --audit-policy-file=/etc/kubernetes/audit-policy.yaml \ 37 --profiling \ 38 --anonymous-auth=false \ 39 --client-ca-file=/etc/kubernetes/cert/ca.pem \ 40 --enable-bootstrap-token-auth=true \ 41 --requestheader-allowed-names="system:metrics-server" \ 42 --requestheader-client-ca-file=/etc/kubernetes/cert/ca.pem \ 43 --requestheader-extra-headers-prefix=X-Remote-Extra- \ 44 --requestheader-group-headers=X-Remote-Group \ 45 --requestheader-username-headers=X-Remote-User \ 46 --service-account-key-file=/etc/kubernetes/cert/ca.pem \ 47 --authorization-mode=Node,RBAC \ 48 --runtime-config=api/all=true \ 49 --enable-admission-plugins=NamespaceLifecycle,LimitRanger,ServiceAccount,DefaultStorageClass,DefaultTolerationSeconds,MutatingAdmissionWebhook,ValidatingAdmissionWebhook,ResourceQuota,NodeRestriction \ 50 --allow-privileged=true \ 51 --apiserver-count=3 \ 52 --event-ttl=168h \ 53 --kubelet-certificate-authority=/etc/kubernetes/cert/ca.pem \ 54 --kubelet-client-certificate=/etc/kubernetes/cert/kubernetes.pem \ 55 --kubelet-client-key=/etc/kubernetes/cert/kubernetes-key.pem \ 56 --kubelet-https=true \ 57 --kubelet-timeout=10s \ 58 --proxy-client-cert-file=/etc/kubernetes/cert/proxy-client.pem \ 59 --proxy-client-key-file=/etc/kubernetes/cert/proxy-client-key.pem \ 60 --service-cluster-ip-range=${SERVICE_CIDR} \ 61 --service-node-port-range=${NODE_PORT_RANGE} \ 62 --logtostderr=true \ 63 --v=2 64 Restart=on-failure 65 RestartSec=10 66 Type=notify 67 LimitNOFILE=65536 68 69 [Install] 70 WantedBy=multi-user.target 71 EOF
提示:本步骤操作仅需要在master01节点操作。
使用 --audit-policy-file 标志将包含策略的文件传递给 kube-apiserver。如果不设置该标志,则不记录事件。
解释:
- --advertise-address:apiserver 对外通告的 IP(kubernetes 服务后端节点 IP);
- --default-*-toleration-seconds:设置节点异常相关的阈值;
- --max-*-requests-inflight:请求相关的最大阈值;
- --etcd-*:访问 etcd 的证书和 etcd 服务器地址;
- --experimental-encryption-provider-config:指定用于加密 etcd 中 secret 的配置;
- --bind-address: https 监听的 IP,不能为 127.0.0.1,否则外界不能访问它的安全端口 6443;
- --secret-port:https 监听端口;
- --insecure-port=0:关闭监听 http 非安全端口(8080);
- --tls-*-file:指定 apiserver 使用的证书、私钥和 CA 文件;
- --audit-*:配置审计策略和审计日志文件相关的参数;
- --client-ca-file:验证 client (kue-controller-manager、kube-scheduler、kubelet、kube-proxy 等)请求所带的证书;
- --enable-bootstrap-token-auth:启用 kubelet bootstrap 的 token 认证;
- --requestheader-*:kube-apiserver 的 aggregator layer 相关的配置参数,proxy-client & HPA 需要使用;
- --requestheader-client-ca-file:用于签名 --proxy-client-cert-file 和 --proxy-client-key-file 指定的证书;在启用了 metric aggregator 时使用;
- --requestheader-allowed-names:不能为空,值为逗号分割的 --proxy-client-cert-file 证书的 CN 名称,这里设置为 "aggregator";
- --service-account-key-file:签名 ServiceAccount Token 的公钥文件,kube-controller-manager 的 --service-account-private-key-file 指定私钥文件,两者配对使用;
- --runtime-config=api/all=true: 启用所有版本的 APIs,如 autoscaling/v2alpha1;
- --authorization-mode=Node,RBAC、--anonymous-auth=false: 开启 Node 和 RBAC 授权模式,拒绝未授权的请求;
- --enable-admission-plugins:启用一些默认关闭的 plugins;
- --allow-privileged:运行执行 privileged 权限的容器;
- --apiserver-count=3:指定 apiserver 实例的数量;
- --event-ttl:指定 events 的保存时间;
- --kubelet-*:如果指定,则使用 https 访问 kubelet APIs;需要为证书对应的用户(上面 kubernetes*.pem 证书的用户为 kubernetes) 用户定义 RBAC 规则,否则访问 kubelet API 时提示未授权;
- --proxy-client-*:apiserver 访问 metrics-server 使用的证书;
- --service-cluster-ip-range: 指定 Service Cluster IP 地址段;
- --service-node-port-range: 指定 NodePort 的端口范围。
提示:如果 kube-apiserver 机器没有运行 kube-proxy,则还需要添加 --enable-aggregator-routing=true 参数。
注意:requestheader-client-ca-file 指定的 CA 证书,必须具有 client auth and server auth;
如果 --requestheader-allowed-names 为空,或者 --proxy-client-cert-file 证书的 CN 名称不在 allowed-names 中,则后续查看 node 或 pods 的 metrics 失败,提示:
1 [root@master01 ~]# kubectl top nodes 2 Error from server (Forbidden): nodes.metrics.k8s.io is forbidden: User "aggregator" cannot list resource "nodes" in API group "metrics.k8s.io" at the cluster scope
5.2 分发systemd
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# for (( i=0; i < 3; i++ )) 4 do 5 sed -e "s/##MASTER_NAME##/${MASTER_NAMES[i]}/" -e "s/##MASTER_IP##/${MASTER_IPS[i]}/" kube-apiserver.service.template > kube-apiserver-${MASTER_IPS[i]}.service 6 done 7 [root@master01 work]# ls kube-apiserver*.service #替换相应的IP 8 9 [root@master01 work]# for master_ip in ${MASTER_IPS[@]} 10 do 11 echo ">>> ${master_ip}" 12 scp kube-apiserver-${master_ip}.service root@${master_ip}:/etc/systemd/system/kube-apiserver.service 13 done #分发systemd
提示:本步骤操作仅需要在master01节点操作。
六 启动并验证
6.1 启动kube-apiserver服务
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# for master_ip in ${MASTER_IPS[@]} 4 do 5 echo ">>> ${master_ip}" 6 ssh root@${master_ip} "mkdir -p ${K8S_DIR}/kube-apiserver" 7 ssh root@${master_ip} "systemctl daemon-reload && systemctl enable kube-apiserver && systemctl restart kube-apiserver" 8 done
提示:本步骤操作仅需要在master01节点操作。
6.2 检查kube-apiserver服务
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# for master_ip in ${MASTER_IPS[@]} 4 do 5 echo ">>> ${master_ip}" 6 ssh root@${master_ip} "systemctl status kube-apiserver |grep 'Active:'" 7 done
提示:本步骤操作仅需要在master01节点操作。
6.3 查看kube-apiserver写入etcd的数据
1 [root@master01 ~]# cd /opt/k8s/work 2 [root@master01 work]# source /root/environment.sh 3 [root@master01 work]# ETCDCTL_API=3 etcdctl 4 --endpoints=${ETCD_ENDPOINTS} 5 --cacert=/opt/k8s/work/ca.pem 6 --cert=/opt/k8s/work/etcd.pem 7 --key=/opt/k8s/work/etcd-key.pem 8 get /registry/ --prefix --keys-only
提示:本步骤操作仅需要在master01节点操作。
6.4 检查集群信息
1 [root@master01 work]# kubectl cluster-info 2 [root@master01 work]# kubectl get all --all-namespaces 3 [root@master01 work]# kubectl get componentstatuses 4 [root@master01 work]# netstat -lnpt|grep 6443 #检查 kube-apiserver 监听的端口
提示:
如果执行 kubectl 命令式时输出如下错误信息,则说明使用的 ~/.kube/config 文件不对,检查《009.Kubernetes二进制master节点部署kubectl》是否正确完成,然后再执行该命令:
The connection to the server localhost:8080 was refused - did you specify the right host or port?
执行 kubectl get componentstatuses 命令时,apiserver 默认向 127.0.0.1 发送请求。目前controller-manager和scheduler还未部署,因此显示Unhealthy;
6443: 接收 https 请求的安全端口,对所有请求做认证和授权;
16443:Nginx反向代理监听端口;
由于关闭了非安全端口,故没有监听 8080。
6.5 授权
授予 kube-apiserver 访问 kubelet API 的权限。
在执行 kubectl exec、run、logs 等命令时,apiserver 会将请求转发到 kubelet 的 https 端口。本实验定义 RBAC 规则,授权 apiserver 使用的证书(kubernetes.pem)用户名(CN:kuberntes)访问 kubelet API 的权限:
1 [root@master01 ~]# kubectl create clusterrolebinding kube-apiserver:kubelet-apis --clusterrole=system:kubelet-api-admin --user kubernetes
提示:本步骤操作仅需要在master01节点操作。