k8s基础维护命令
文章目录
基础维护
# 当需要对主机进行维护升级时，首先将节点主机设置成不可调度模式：
kubectl cordon［nodeid］
[root@ziji-k8s-master1-192-168-1-16 coredns]# kubectl cordon 192.168.1.17 node/192.168.1.17 cordoned [root@ziji-k8s-master1-192-168-1-16 coredns]# kubectl get node NAME STATUS ROLES AGE VERSION 192.168.1.16 Ready,SchedulingDisabled master 25d v1.11.4 192.168.1.17 Ready,SchedulingDisabled node 27d v1.11.4 192.168.1.18 Ready node 27d v1.11.4 192.168.1.19 Ready node 27d v1.11.4 192.168.1.20 Ready dc1-db,node 15d v1.11.4
# 然后需要将主机上正在运行的容器驱赶到其它可用节点：
kubectl drain ［nodeid］
# 给予900秒宽限期优雅的调度
kubectl drain node1.k8s.novalocal --grace-period=120
# 当容器迁移完毕后，运维人员可以对该主机进行操作，配置升级性能参数调优等等。当对主机的维护操作完毕后， 再将主机设置成可调度模式：
kubectl uncordon [nodeid]
#强制删除已经Terminating pod
kubectl get pod -n zk-kafka kubectl delete zookeeper-2-4n426 -n zk-kafka --grace-period=0 --force
删除Evicted （被驱逐的pod）
kubectl get pods -n apache-test | grep Evicted | awk '{print $1}' | xargs kubectl delete pod -n apache-test
更新操作
注意：当且仅当部署的pod模板（即.spec.template）更改时，才会触发Deployment的部署，例如，如果更新模板的标签或容器图像。其他更新（例如扩展部署）不会触发部署。
kubectl set image deployment/web-service web-service=web-service:v1.9.1 --record deployment "web-service" image updated
也可以进行编辑进行更改
$ kubectl edit deployment/web-service -n ziji deployment "nginx-deployment" edited
回滚操作< deployment|daemonset >
查看版本发布历史
kubectl rollout history deployment/web-service -n ziji deployments "web-service" REVISION CHANGE-CAUSE 23 <none> 27 <none> 28 <none>
回滚到上一个版本
kubectl rollout undo deployment -n ziji web-service
#查看deployment对象信息
kubectl describe deployment -n ziji web-service Name: web-service Namespace: ziji CreationTimestamp: Sun, 02 Dec 2018 18:31:43 +0800 Labels: app=apache-v1 name=web-service Annotations: deployment.kubernetes.io/revision=28
#查看replicasets对象信息
kubectl get rs -o wide -n ziji
回滚到指定版本
kubectl rollout undo deployment/web-service --to-revision=27
查看发布的状态
kubectl rollout status deployment/web-service -n ziji deployment "web-service" successfully rolled out
扩容操作
设置自动扩容（水平拉伸）, 当pod cpu使用率超过10%, 最小创建pod数为1， 最大3
kubectl autoscale deployment -n ziji web-service --min=1 --max=3 --cpu-percent=10
指定扩容pod副本数量
kubectl scale deployment -n ziji web-service --replicas 3
滚动更新
spec: replicas: 10 selector: matchLabels: service: http-server strategy: type: RollingUpdate rollingUpdate: maxSurge: 1 maxUnavailable: 1 minReadySeconds: 5
暂停滚动更新
kubectl rollout pause deployment <deployment>
继续滚动更新
kubectl rollout resume deployment <deployment>
保留版本次数
... spec: replicas: 10 selector: matchLabels: service: http-server strategy: type: RollingUpdate rollingUpdate: maxSurge: 1 maxUnavailable: 1 minReadySeconds: 5 revisionHistoryLimit: 10 ...
常用命令
- 创建：kubectl run <name> –image=<image> 或者 kubectl create -f manifest.yaml
- 查询：kubectl get <resource>
- 更新 kubectl set 或者 kubectl patch
- 删除：kubectl delete <resource> <name> 或者 kubectl delete -f manifest.yaml
- 查询 Pod IP：kubectl get pod <pod-name> -o jsonpath='{.status.podIP}’
- 容器内执行命令：kubectl exec -ti <pod-name> sh
- 容器日志：kubectl logs [-f] <pod-name>
- 导出服务：kubectl expose deploy <name> –port=80
- Base64 解码：
kubectl get secret SECRET -o go-template='{{ .data.KEY | base64decode }}'
查看endpoints
[root@feiba-k8s-master1-192-168-1-16 ~]# kubectl get ep -n ziji-test NAME ENDPOINTS AGE glusterfs-dynamic-pvc-zup-config 192.168.1.12:1,192.168.1.13:1,192.168.1.14:1 + 1 more... 34d glusterfs-dynamic-pvc-zup-gateway 192.168.1.12:1,192.168.1.13:1,192.168.1.14:1 + 1 more... 34d glusterfs-dynamic-pvc-zup-payment-web 192.168.1.12:1,192.168.1.13:1,192.168.1.14:1 + 1 more... 32d glusterfs-dynamic-pvc-zup-task 192.168.1.12:1,192.168.1.13:1,192.168.1.14:1 + 1 more... 33d games 10.96.20.3:8099 4d gateway 10.96.20.10:6674 18d gateway-admin 10.96.66.7:5674 4d merchant 10.96.20.8:8080 33d mining 10.96.66.5:8085 33d payment-web 10.96.20.13:8084 32d register 10.96.20.7:8888 18d task 10.96.20.6:8083 33d
[root@ziji-k8s-master1-192-168-1-16 ~]# kubectl get endpoints mining -n test NAME ENDPOINTS AGE mining 10.96.66.5:8085 33d
日志查看
kubectl logs nginx kubectl logs -f -c ruby web-1
连接到一个正在运行的容器
kubectl attach 用于连接到一个正在运行的容器
文件拷贝
kubectl cp /tmp/foo_dir <some-pod>:/tmp/bar_dir #拷贝文件到容器中 kubectl cp <some-namespace>/<some-pod>:/tmp/foo /tmp/bar #从容器拷贝文件到本地
反亲和性
affinity: podAntiAffinity: requiredDuringSchedulingIgnoredDuringExecution: - labelSelector: matchExpressions: - key: app operator: In values: - <APP_NAME> topologyKey: "kubernetes.io/hostname"
