本文档介绍部署高可用 kube-controller-manager 集群的步骤。
该集群包含 3 个节点,启动后将通过竞争选举机制产生一个 leader 节点,其它节点为阻塞状态。当 leader 节点不可用后,剩余节点将再次进行选举产生新的 leader 节点,从而保证服务的可用性。
本文档介绍部署高可用 kube-controller-manager 集群的步骤。
该集群包含 3 个节点,启动后将通过竞争选举机制产生一个 leader 节点,其它节点为阻塞状态。当 leader 节点不可用后,剩余节点将再次进行选举产生新的 leader 节点,从而保证服务的可用性。
为保证通信安全,本文档先生成 x509 证书和私钥,kube-controller-manager 在如下两种情况下使用该证书:
1)与 kube-apiserver 的安全端口通信时;
2)在安全端口 (https,10252) 输出 prometheus 格式的 metrics;
配置之前需要先安装 kubelet,flannel 等组件,不过前边已经安装,现在直接进入配置。
1、创建 kube-controller-manager 证书和私钥
创建证书签名请求:
cat > /data/cert/kube-controller-manager-csr.json <
- hosts 列表包含所有 kube-controller-manager 节点 IP;
- CN 为 system:kube-controller-manager、O 为 system:kube-controller-manager,kubernetes 内置的 ClusterRoleBindings system:kube-controller-manager 赋予 kube-controller-manager 工作所需的权限
生成证书和私钥:
[root@k8s-master data]#
cfssl gencert -ca=/etc/kubernetes/cert/ca.pem \
-ca-key=/etc/kubernetes/cert/ca-key.pem \
-config=/etc/kubernetes/cert/ca-config.json \
-profile=kubernetes kube-controller-manager-csr.json | cfssljson -bare kube-controller-manager
将生成的证书和私钥分发到所有 master 节点:
cat > /data/script/magic41_distribute_cert_All_service_.sh << "EOF"
#!/bin/bash
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
scp /data/cert/kube-controller-manager*.pem root@${node_ip}:/etc/kubernetes/cert/
done
EOF
2、创建和分发 kubeconfig 文件
kubeconfig 文件包含访问 apiserver 的所有信息,如 apiserver 地址、CA 证书和自身使用的证书;
[root@k8s-master data]# source /opt/k8s/bin/environment.sh
kubectl config set-cluster kubernetes \
--certificate-authority=/etc/kubernetes/cert/ca.pem \
--embed-certs=true \
--server=${KUBE_APISERVER} \
--kubeconfig=kube-controller-manager.kubeconfig
kubectl config set-credentials system:kube-controller-manager \
--client-certificate=kube-controller-manager.pem \
--client-key=kube-controller-manager-key.pem \
--embed-certs=true \
--kubeconfig=kube-controller-manager.kubeconfig
kubectl config set-context system:kube-controller-manager \
--cluster=kubernetes \
--user=system:kube-controller-manager \
--kubeconfig=kube-controller-manager.kubeconfig
kubectl config use-context system:kube-controller-manager --kubeconfig=kube-controller-manager.kubeconfig
分发 kubeconfig 到所有 master 节点:
cat > /data/script/magic42_distribute_kubeconfig_all_Node_service.sh << "EOF"
#!/bin/bash
# 分发 kubeconfig 到所有 master 节点
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
scp /data/scriptable/kubeconfig/kube-controller-manager.kubeconfig root@${node_ip}:/etc/kubernetes/
done
EOF
3、创建和分发 kube-controller-manager systemd unit 文件
[root@k8s-master data]# source /opt/k8s/bin/environment.sh
cat > /data/service/kube-controller-manager.service <
- --port=0:关闭监听 http /metrics 的请求,同时 --address 参数无效,--bind-address 参数有效;
- --secure-port=10252、--bind-address=0.0.0.0: 在所有网络接口监听 10252 端口的 https /metrics 请求;
- --kubeconfig:指定 kubeconfig 文件路径,kube-controller-manager 使用它连接和验证 kube-apiserver;
- --cluster-signing-*-file:签名 TLS Bootstrap 创建的证书;
- --cluster-signing-duration:指定 TLS Bootstrap 证书的有效期;
- --root-ca-file:放置到容器 ServiceAccount 中的 CA 证书,用来对 kube-apiserver 的证书进行校验;
- --service-account-private-key-file:签名 ServiceAccount 中 Token 的私钥文件,必须和 kube-apiserver 的 --service-account-key-file 指定的公钥文件配对使用;
- --service-cluster-ip-range :指定 Service Cluster IP 网段,必须和 kube-apiserver 中的同名参数一致;
- --leader-elect=true:集群运行模式,启用选举功能;被选为 leader 的节点负责处理工作,其它节点为阻塞状态;
- --feature-gates=RotateKubeletServerCertificate=true:开启 kublet server 证书的自动更新特性;
- --controllers=*,bootstrapsigner,tokencleaner:启用的控制器列表,tokencleaner 用于自动清理过期的 Bootstrap token;
- --horizontal-pod-autoscaler-*:custom metrics 相关参数,支持 autoscaling/v2alpha1;
- --tls-cert-file、--tls-private-key-file:使用 https 输出 metrics 时使用的 Server 证书和秘钥;
- --use-service-account-credentials=true:
- User=k8s:使用 k8s 账户运行;
kube-controller-manager 不对请求 https metrics 的 Client 证书进行校验,故不需要指定 –tls-ca-file 参数,而且该参数已被淘汰
分发 systemd unit 文件到所有 master 节点:
cat > /data/script/magic43_distribute_all_Node_servier.sh << "EOF"
#!/bin/bash
# 分发 systemd unit 文件到所有 master 节点
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
scp /data/service/kube-controller-manager.service root@${node_ip}:/etc/systemd/system/
done
EOF
4、kube-controller-manager 的权限
ClusteRole: system:kube-controller-manager 的权限很小,只能创建 secret、serviceaccount 等资源对象,各 controller 的权限分散到 ClusterRole system:controller:XXX 中。
需要在 kube-controller-manager 的启动参数中添加 、–use-service-account-credentials=true 参数,这样 main controller 会为各 controller 创建对应的 ServiceAccount XXX-controller。
内置的 ClusterRoleBinding system:controller:XXX 将赋予各 XXX-controller ServiceAccount 对应的 ClusterRole system:controller:XXX 权限
5、启动 kube-controller-manager 服务
cat > /data/script/magic44_start_kube-controller-manager_service.sh << "EOF"
#!/bin/bash
# 启动 kube-controller-manager 服务
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh root@${node_ip} "mkdir -p /var/log/kubernetes && chown -R k8s /var/log/kubernetes"
ssh root@${node_ip} "systemctl daemon-reload && systemctl enable kube-controller-manager && systemctl start kube-controller-manager"
done
EOF
6、检查服务运行状态
cat > /data/script/magic45_check_kube-controller-manager_service.sh << "EOF"
#!/bin/bash
# 检查服务运行状态
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh root@${node_ip} "systemctl status kube-controller-manager|grep Active"
done
EOF
输出如下内容:
[root@k8s-master scriptable]# bash magic.sh
>>> 172.68.96.101
Active: active (running) since Fri XXXXX CST; Xh ago
>>> 172.68.96.102
Active: active (running) since Fri XXXXX CST; Xh ago
>>> 172.68.96.103
Active: active (running) since Fri XXXXX CST; Xh ago
则为正常,否则,查看日志:
journalctl -xu kube-controller-manager
7、查看输出的 metric
注意:以下命令在 kube-controller-manager 节点上执行
kube-controller-manager 监听 10252 端口,接收 https 请求:
[root@k8s-master data]# sudo netstat -lnpt|grep kube-controll
tcp 0 0 127.0.0.1:10252 0.0.0.0:* LISTEN 14858/kube-controll
[root@k8s-master data]# curl -s --cacert /etc/kubernetes/cert/ca.pem https://127.0.0.1:10252/metrics |head
# HELP ClusterRoleAggregator_adds Total number of adds handled by workqueue: ClusterRoleAggregator
# TYPE ClusterRoleAggregator_adds counter
ClusterRoleAggregator_adds 17
# HELP ClusterRoleAggregator_depth Current depth of workqueue: ClusterRoleAggregator
# TYPE ClusterRoleAggregator_depth gauge
ClusterRoleAggregator_depth 0
# HELP ClusterRoleAggregator_queue_latency How long an item stays in workqueueClusterRoleAggregator before being requested.
# TYPE ClusterRoleAggregator_queue_latency summary
ClusterRoleAggregator_queue_latency{quantile="0.5"} NaN
ClusterRoleAggregator_queue_latency{quantile="0.9"} NaN
- curl --cacert CA 证书用来验证 kube-controller-manager https server 证书;
8、查看当前的 leader
[root@k8s-master data]# kubectl get endpoints kube-controller-manager --namespace=kube-system -o yaml
apiVersion: v1
kind: Endpoints
metadata:
annotations:
control-plane.alpha.kubernetes.io/leader: '{"holderIdentity":"kube-node1_XXXXXXXXXXXXXXX","leaseDurationSeconds":15,"acquireTime":"XXXXXXXXXXXXXXX","renewTime":"XXXXXXXXXXXXXXX","leaderTransitions":0}'
creationTimestamp: XXXX-XX-XX
name: kube-controller-manager
namespace: kube-system
resourceVersion: "29655"
selfLink: /api/v1/namespaces/kube-system/endpoints/kube-controller-manager
uid: XXXXXXXXXXXXXXX
可见,当前的 leader 为 kube-node1 节点
9、测试 kube-controller-manager 集群的高可用
停掉一个或两个节点的 kube-controller-manager 服务,观察其它节点的日志,看是否获取了 leader 权限。
现在停掉 kube-node1 上的 kube-controller-manager
[root@k8s-master data]# systemctl stop kube-controller-manager
[root@k8s-master data]# systemctl status kube-controller-manager |grep Active
Active: inactive (dead) since Sat XXXXXXXXX CST; XXs ago
再查看一下当前的 leader:
[root@k8s-master data]# kubectl get endpoints kube-controller-manager --namespace=kube-system -o yaml
apiVersion: v1
kind: Endpoints
metadata:
annotations:
control-plane.alpha.kubernetes.io/leader: '{"holderIdentity":"XXXXXXXXX4","leaseDurationSeconds":15,"acquireTime":"kube-node3_XXXXXXXXX","renewTime":"XXXXXXXXX","leaderTransitions":1}'
creationTimestamp: XXXXXXXXX
name: kube-controller-manager
namespace: kube-system
resourceVersion: "29971"
selfLink: /api/v1/namespaces/kube-system/endpoints/kube-controller-manager
uid: XXXXXXXXX
可以看到已经自动漂移到 kube-node3 上去了
10,参考资料
1)关于 controller 权限和 use-service-account-credentials 参数:https://github.com/kubernetes/kubernetes/issues/48208
2)kublet 认证和授权:https://kubernetes.io/docs/admin/kubelet-authentication-authorization/#kubelet-authorization