【精选】kubernetes工作负载控制器的应用 |
您所在的位置：网站首页 › k8s工作负载的重载 › 【精选】kubernetes工作负载控制器的应用 |
【精选】kubernetes工作负载控制器的应用
|
kubernetes工作负载控制器的应用
工作负载控制器是什么Deployment介绍deployment的资源清单文件Deployment 滚动升级Deployment 水平扩容Deployment 回滚Deployment 删除Deployment ReplicaSet
DameonSetJob 和 CronJob
工作负载控制器是什么
工作负载控制器（Workload Controllers）是K8s的一个抽象概念，用于更高级层次对象，部署和管理Pod 常用的工作负载控制器： Deployment ：无状态应用部署StatefulSet ：有状态应用部署DaemonSet ：确保所有Node运行同一个PodJob ：一次性任务Cronjob ：定时任务控制器的作用： 管理Pod对象使用标签与Pod关联控制器实现了Pod的运维，例如滚动更新、伸缩、副本管理、维护Pod状态等示例：
为了更好地解决服务编排的问题，k8s在V1.2版本开始，引入了deployment控制器，值得一提的是，这种控制器并不直接管理pod， 而是通过管理replicaset来间接管理pod，即：deployment管理replicaset，replicaset管理pod。所以deployment比replicaset的功能更强大。 deployment的主要功能有下面几个： 支持replicaset的所有功能支持发布的停止、继续支持版本的滚动更新和版本回退 deployment的资源清单文件 apiVersion: apps/v1 #版本号 kind: Deployment #类型 metadata: #元数据 name: #rs名称 namespace: #所属命名空间 labels: #标签 controller: deploy spec: #详情描述 replicas: #副本数量 revisionHistoryLimit: #保留历史版本，默认是10 paused: #暂停部署，默认是false progressDeadlineSeconds: #部署超时时间(s)，默认是600 strategy: #策略 type: RollingUpdates #滚动更新策略 rollingUpdate: #滚动更新 maxSurge: #最大额外可以存在的副本数，可以为百分比，也可以为整数 maxUnavaliable: #最大不可用状态的pod的最大值，可以为百分比，也可以为整数 selector: #选择器，通过它指定该控制器管理哪些pod matchLabels: #Labels匹配规则 app: nginx-pod matchExpressions: #Expression匹配规则 - {key: app, operator: In, values: [nginx-pod]} template: #模板，当副本数量不足时，会根据下面的模板创建pod副本 metadata: labels: app: nginx-pod spec: containers: - name: nginx image: nginx:1.17.1 ports: - containerPort: 80 [root@master ~]# cat test.yaml --- apiVersion: apps/v1 //api的版本信息 kind: Deployment //类型 metadata: name: test //容器名称 namespace: default //使用默认的名称空间 spec: replicas: 3 //三个容器副本 selector: matchLabels: app: busybox //容器标签 template: metadata: labels: app: busybox spec: containers: - name: b1 image: busybox //使用的镜像 command: ["/bin/sh","-c","sleep 9000"] [root@master ~]# kubectl apply -f test.yaml deployment.apps/amu created [root@master ~]# kubectl get pod NAME READY STATUS RESTARTS AGE test-864fd4fb54-4hz2h 1/1 Running 0 1ms test-864fd4fb54-j9ht4 1/1 Running 0 1m test-864fd4fb54-w485p 1/1 Running 0 1m Deployment 滚动升级kubectl apply -f xxx.yaml kubectl set image deployment/web nginx=nignx:1.16 kubectl edit deployment/web 滚动升级：K8s对Pod升级的默认策略，通过使用新版本Pod逐步更新旧版本Pod，实现停机发布，用户无感知 // 滚动更新策略 spec: replicas: 3 revisionHistoryLimit: 10 selector: matchLabels: app: web strategy: rollingUpdate: maxSurge: 25% maxUnavaliable: 25% type: RollingUpdate maxSurge：滚动更新过程中最大Pod副本数，确保在更新时启动的Pod数量比期望（replicsa）Pod数量最大多出25%maxUnavailable：滚动更新过程中最大不可用Pod副本数，确保在更新时最大25%Pod数量不可用，即确保75%Pod数量是可用状态示例： // 创建四个httpd容器 [root@master ~]# cat test1.yaml --- apiVersion: apps/v1 kind: Deployment metadata: name: web namespace: default spec: replicas: 4 selector: matchLabels: app: httpd template: metadata: labels: app: web spec: containers: - name: httpd image: xxz-1/httpd:v0.2 imagePullPolicy: IfNotPresent [root@master ~]# kubectl apply -f test1.yaml deployment.apps/web created [root@master ~]# kubectl get pods web-5c688b9779-7gi6t 1/1 Running 0 50s web-5c688b9779-9unfy 1/1 Running 0 50s web-5c688b9779-ft69k 1/1 Running 0 50s web-5c688b9779-vmlkg 1/1 Running 0 50s [root@master ~]# cat test1.yaml --- apiVersion: apps/v1 kind: Deployment metadata: name: web namespace: default spec: replicas: 4 strategy: rollingUpdate: // 滚动更新 maxSurge: 25% // 最大可超出25% maxUnavailable： 25% // 最多不能使用25% type: RollingUpdate selector: matchLabels: app: httpd template: metadata: labels: app: httpd spec: containers: - name: httpd image: dockerimages123/httpd:v0.2 imagePullPolicy: IfNotPresent [root@master ~]# kubectl apply -f test1.yaml deployment.apps/web configured // 暂未发生变化 [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE web-5c697b9779-7gi6t 1/1 Running 0 8m web-5c697b9779-9unfy 1/1 Running 0 8m web-5c697b9779-ft69k 1/1 Running 0 8m web-5c697b9779-vmlkg 1/1 Running 0 8m // 将镜像修改为httpd，image=httpd // 再次应用 [root@master ~]# kubectl apply -f test1.yaml deployment.apps/web configured // 我们发现删除三个，启动了两个新的和一个旧的web [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE web-5d688b9779-7gi6t 1/1 Terminating 0 18m web-5d688b9779-9unfy 0/1 Running 0 20m web-5d688b9779-ft69k 0/1 Terminating 0 20m web-5d688b9779-vmlkg 0/1 Terminating 0 20m web-f8bcfc88-vddfk 0/1 Running 0 80s web-f8bcfc88-yur8y 0/1 Running 0 80s web-f8bcfc88-t9ryx 1/1 ContainerCreating 0 55s web-f8bcfc88-k07k 1/1 ContainerCreating 0 56s // 最后变为4个新 [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE web-f8bcfc88vddfk 1/1 Running 0 80s web-f8bcfc88-yur8y 1/1 Running 0 80s web-f8bcfc88-t9ryx 1/1 Running 0 55s web-f8bcfc88-k07k 1/1 Running 0 56s Deployment 水平扩容修改yanl里replicas值，再apply kubectl scale deployment web --replicas=10 [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE web-f8bcfc88vddfk 1/1 Running 0 80s web-f8bcfc88-yur8y 1/1 Running 0 80s web-f8bcfc88-t9ryx 1/1 Running 0 55s web-f8bcfc88-k07k 1/1 Running 0 56s // 创建10个容器 [root@master ~]# cat test1.yaml --- apiVersion: apps/v1 kind: Deployment metadata: name: web namespace: default spec: replicas: 10 strategy: rollingUpdate: maxSurge: 55% maxUnavailable: 50% type: RollingUpdate selector: matchLabels: app: httpd template: metadata: labels: app: httpd spec: containers: - name: httpd image: dockerimages123/httpd:v0.2 imagePullPolicy: IfNotPresent [root@master ~]# kubectl apply -f test1.yaml deployment.apps/web created [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE web-5c688b9779-pb4x8 1/1 Running 0 50s web-5c688b9779-kf8vq 1/1 Running 0 50s web-5c688b9779-ki8s3 1/1 Running 0 50s web-5c688b9779-o9gx6 1/1 Running 0 50s web-5c688b9779-i8g4w 1/1 Running 0 50s web-5c688b9779-olgxt 1/1 Running 0 50s web-5c688b9779-khctw 1/1 Running 0 50s web-5c688b9779-ki8d6 1/1 Running 0 50s web-5c688b9779-i9g5s 1/1 Running 0 50s web-5c688b9779-jsj8k 1/1 Running 0 50s // 修改replicas值，实现水平扩缩容 [root@master ~]# vim test1.yaml replicas: 3 [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE web-5c688b9779-pb4x8 1/1 Running 0 50s web-5c688b9779-kf8vq 1/1 Running 0 50s web-5c688b9779-ki8s3 1/1 Running 0 50s web-5c688b9779-o9gx6 1/1 Running 0 50s web-5c688b9779-i8g4w 1/1 Running 0 50s web-5c688b9779-olgxt 1/1 Running 0 50s web-5c688b9779-khctw 1/1 Running 0 50s web-5c688b9779-ki8d6 1/1 Running 0 50s web-5c688b9779-i9g5s 1/1 Running 0 50s web-5c688b9779-jsj8k 1/1 Running 0 50s web-i9olh676-jdkrg 0/1 ContainerCreating 0 8s web-i9olh676-opy5b 0/1 ContainerCreating 0 8s web-i9olh676-k8st4 0/1 ContainerCreating 0 8s [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE web-i9olh676-jdkrg 1/1 Running 0 2m19s web-i9olh676-opy5b 1/1 Running 0 2m19s web-i9olh676-k8st4 1/1 Running 0 2m19s Deployment 回滚kubectl rollout history deployment/web // 查看历史发布版本 kubectl rollout undo deployment/web // 回滚上一个版本 kubectl rollout undo deployment/web --to-revision=2 // 回滚历史指定版本 [root@master ~]# kubectl rollout history deploy/web deployment.apps/web REVISION CHANGE-CAUSE 1 2 // 如果你回滚的是当前版本，就会跳过 [root@master ~]# kubectl rollout undo deploy/web --to-revision 2 deployment.apps/web skipped rollback (current template already matches revision 2) // 回滚第一个版本 [root@master ~]# kubectl rollout undo deploy/web --to-revision 1 deployment.apps/web rolled back // 回滚成功 [root@master ~]# kubectl rollout history deploy/web deployment.apps/web REVISION CHANGE-CAUSE 2 3 Deployment 删除kubectl delete deploy/web kubectl delete svc/web kubectl delete pods/web // 创建 [root@master ~]# kubectl create deployment nginx --image=nginx deployment.apps/nginx created [root@master ~]# kubectl get deploy NAME READY UP-TO-DATE AVAILABLE AGE nginx 1/1 1 1 35s web 3/3 3 3 20m // 删除一个pod [root@master ~]# kubectl delete deploy/nginx deployment.apps "nginx" deleted [root@master ~]# kubectl get deploy NAME READY UP-TO-DATE AVAILABLE AGE web 3/3 3 3 20m // 删除所有 [root@master ~]# kubectl delete deployment --all deployment.apps "web" deleted [root@master ~]# kubectl get deployment No resources found in default namespace. Deployment ReplicaSetPod副本数量管理，不断对比当前Pod数量与期望Pod数量 Deployment每次发布都会创建一个RS作为记录，用于实现回滚 [root@master ~]# kubectl get rs NAME DESIRED CURRENT READY AGE web-5d7hy50s8a 3 3 3 6m4s web-f8bki8h5 0 0 0 6m25s kubectl rollout history deployment web // 版本对应RS记录 [root@master ~]# kubectl rollout history deployment web deployment.apps/web REVISION CHANGE-CAUSE 2 3 [root@master ~]# kubectl get rs NAME DESIRED CURRENT READY AGE web-5d7hy50s8a 3 3 3 6m4s web-f8bki8h5 0 0 0 6m25s // 将镜像换为httpd，其他变 [root@master ~]# kubectl apply -f test2.yaml //再次应用 deployment.apps/web configured [root@master ~]# kubectl get pods web-5d688b9745-dpmsd 1/1 Terminating 0 9m web-5d688b9745-q6dls 1/1 Terminating 0 9m web-i80gjk6t-ku6f4 0/1 ContainerCreating 0 5s web-i80gjk6t-9j5tu 1/1 Running 0 5s web-i80gjk6t-9ir4h 1/1 Running 0 5s [root@master ~]# kubectl get rs NAME DESIRED CURRENT READY AGE web-5d7hy50s8a 0 0 0 12m web-f8bki8h5 3 3 3 12mReplicaSet：由Deployment里面的relicas的参数控制的 [root@node2 ~]# docker ps | grep web 4c938ouhc8j0 dabutse0c4fy "httpd-foreground" 13 seconds ago Up 12 seconds k8s_httpd_web-f8bcfc88-4rkfx_default_562616cd-1552-4610-bf98-e470225e4c31_1 452713eeccad registry.aliyuncs.com/google_containers/pause:3.6 "/pause" 5 minutes ago Up 5 minutes k8s_POD_web-f8bcfc88-4rkfx_default_562616cd-1552-4610-bf98-e470225e4c31_0 // 删除一个 [root@node2 ~]# docker kill 4c938ad0c01d 8htw4ad0cu8s [root@master ~]# kubectl get rs NAME DESIRED CURRENT READY AGE web-5d7hy50s8a 0 0 0 15m web-f8bki8h5 3 3 3 15m [root@master ~]# kubectl get rs NAME DESIRED CURRENT READY AGE web-5d7hy50s8a 0 0 0 16m web-f8bki8h5 3 3 3 18m [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE web-o96gb3sm-9ht4c 1/1 Running 2 (80s ago) 6m32s web-o96gb3sm-ki85s 1/1 Running 0 6m32s web-o96gb3sm-ku5sg 1/1 Running 0 6m32s DameonSet在每一个Node上运行一个Pod 新加入的Node也同样会自动运行一个Pod // 删除资源，容器也会被删除 [root@master ~]# kubectl delete -f test1.yml deployment.apps "web" deleted [root@master ~]# cat daemon.yml --- apiVersion: apps/v1 kind: DaemonSet // 类型是DaemonSet metadata: name: filebeat namespace: kube-system spec: selector: matchLabels: name: filebeat template: metadata: labels: name: filebeat spec: containers: // 记录日志的镜像 - name: log image: elastic/filebeat:7.16.2 imagePullPolicy: IfNotPresent [root@master ~]# kubectl get pods -n kube-system NAME READY STATUS RESTARTS AGE coredns-6d8c4cb4d-9m5jg 1/1 Running 14 (4h36m ago) 6d4h coredns-6d8c4cb4d-mp662 1/1 Running 14 (4h36m ago) 6d4h etcd-master 1/1 Running 13 (3h30m ago) 6d4h kube-apiserver-master 1/1 Running 13 (3h30m ago) 6d4h kube-controller-manager-master 1/1 Running 14 (3h30m ago) 6d4h kube-flannel-ds-g9jsh 1/1 Running 9 (3h30m ago) 6d1h kube-flannel-ds-qztxc 1/1 Running 9 (3h30m ago) 6d1h kube-flannel-ds-t8lts 1/1 Running 13 (3h30m ago) 6d1h kube-proxy-q2jmh 1/1 Running 12 (3h30mago) 6d4h kube-proxy-r28dn 1/1 Running 13 (3h30m ago) 6d4h kube-proxy-x4cns 1/1 Running 13 (3h30m ago) 6d4h kube-scheduler-master 1/1 Running 15 (3h30m ago) 6d4h [root@master ~]# kubectl apply -f daemon.yaml deployment.apps/filebeat created [root@master ~]# kubectl get pods -n kube-system NAME READY STATUS RESTARTS AGE coredns-6d8c4cb4d-9m5jg 1/1 Running 14 (4h36m ago) 6d4h coredns-6d8c4cb4d-mp662 1/1 Running 14 (4h36m ago) 6d4h etcd-master 1/1 Running 13 (3h30m ago) 6d4h kube-apiserver-master 1/1 Running 13 (3h30m ago) 6d4h kube-controller-manager-master 1/1 Running 14 (3h30m ago) 6d4h kube-flannel-ds-g9jsh 1/1 Running 9 (3h30m ago) 6d1h kube-flannel-ds-qztxc 1/1 Running 9 (3h30m ago) 6d1h kube-flannel-ds-t8lts 1/1 Running 13 (3h30m ago) 6d1h kube-proxy-q2jmh 1/1 Running 12 (3h30mago) 6d4h kube-proxy-r28dn 1/1 Running 13 (3h30m ago) 6d4h kube-proxy-x4cns 1/1 Running 13 (3h30m ago) 6d4h kube-scheduler-master 1/1 Running 15 (3h30m ago) 6d4h filebeat-oiugt 1/1 Running 0 50s filebeat-9jhgt 1/1 Running 0 50s [root@master ~]# kubectl get pods -n kube-system -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES filebeat-9ck6z 1/1 Running 0 3m9s 10.242.2.215 node2.example.com filebeat-d2psf 1/1 Running 0 3m9s 10.242.1.161 node1.example.com Job 和 CronJobJob分为普通任务（Job）和定时任务（CronJob） 一次性执行(Job) 应用场景：离线数据处理，视频解码等业务定时任务(CronJob) 应用场景：通知，备份 [root@master ~]# cat job.yaml apiVersion: batch/v1 kind: Job metadata: name: test spec: template: spec: containers: - name: test image: perl command: ["perl","-Mbignum=bpi","-wle","print bpi(2000)"] restartPolicy: Never backoffLimit: 4 [root@master ~]# kubectl apply -f job.yaml job.batch/test created [root@master ~]# kubectl get pods -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES test-27xrt 0/1 ContainerCreating 0 10m28s node2.example.com [root@node2 ~]# docker ps | grep test e55ac8842c89 registry.aliyuncs.com/google_containers/pause:3.6 "/pause" 6 minutes ago Up 6 minutes k8s_POD_test-27xrt_default_698b4c91-ef54-4fe9-b62b-e0abcujhts5o90_9 [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE test-27xrt 0/1 Completed 0 15m [root@node2 ~]# docker images | grep perl perl latest f9596eddf06f 5 days ago 568MB [root@master ~]# kubectl describe job/test Pods Statuses: 0 Active / 1 Succeeded / 0 Failed [root@master ~]# kubelet --version Kubernetes v1.23.1 [root@master ~]# cat cronjob.yaml --- apiVersion: batch/v1 kind: CronJob metadata: name: hello spec: schedule: "*/1****" jobTemplate: spec: template: spec: containers: - name: hello image: busybox args: - /bin/sh - -c - date;echo Hello world restartPolicy: OnFailure [root@master ~]# kubectl apply -f cronjob.yaml Warning: batch/v1beta1 CronJob is deprecated in v1.21+, unavailable in v1.25+; use batch/v1 CronJob [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE hello-kihtwoab-kox6w 0/1 Completed 0 5m42s hello-kihtwoab-o96vw 0/1 Completed 0 90s hello-kihtwoab-kus6n 0/1 Completed 0 76s [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE hello-kihtwoab-kox6w 0/1 Completed 0 6m26s hello-kihtwoab-o96vw 0/1 Completed 0 2m11s hello-kihtwoab-kus6n 0/1 Completed 0 2m10s hello-kuhdoehs-ki8gr 0/1 ContainerCreating 0 36s [root@master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE hello-kihtwoab-o96vw 0/1 Completed 0 2m11s hello-kihtwoab-kus6n 0/1 Completed 0 2m10s hello-kuhdoehs-ki8gr 0/1 Completed 0 45s [root@master ~]# kubectl describe cronjob/hello Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal SuccessfulCreate 5m36s (x133 over 135m) cronjob-controller (combined from similar events): Created job hello-kisgejxbw |
【本文地址】
今日新闻 |
推荐新闻 |