kubenetes中的pod删除策略 级联删除与非级联删除

kubenetes中的pod删除策略 级联删除与非级联删除,第1张

StatefulSet 有状态应用[有状态应用] 有状态:StatefulSet   - 集群节点之间的关系。   - 数据不完全一致。   - 实例之间不对等的关系。   - 依靠外部存储的应用。   - 通过dns维持身份   - 每个pod都有特定的名称和网络标识(如pod名是由statefulSet名+有序的数字组成(0、1、2..))      redis是有状态应用 StatefulSet(有状态集,缩写为sts)常用于部署有状态的且需要有序启动的应用程序,比如在进行SpringCloud项目容器化时,Eureka的部署是比较适合用StatefulSet部署方式的,可以给每个Eureka实例创建一个唯一且固定的标识符,并且每个Eureka实例无需配置多余的Service,其余Spring Boot应用可以直接通过Eureka的Headless Service即可进行注册。 Eureka的statefulset的资源名称是 eureka,eureka-0 eureka-1 eureka-2 Service:headless service,没有ClusterIP [每个] eureka-svc Eureka-0.eureka-svc.NAMESPACE_NAME eureka-1.eureka-svc … 连接 eureka的资源名称为:eureka statefullset示例 apiVersion: v1 kind: Service metadata: name: nginx labels: app: nginx spec: ports: - port: 80 name: web clusterIP: None selector: app: nginx --- apiVersion: apps/v1 kind: StatefulSet metadata: name: web spec: serviceName: "nginx" replicas: 2 selector: matchLabels: app: nginx template: metadata: labels: app: nginx spec: containers: - name: nginx image: nginx ports: - containerPort: 80 name: web [root@k8s-master01 statefullset]# kubectl create -f statefullset.yaml service/nginx created statefulset.apps/web created #扩容: replicas: 2 #[定义副本集数量] [root@k8s-master01 statefullset]# kubectl get pod web-0 1/1 Running 0 1m web-1 1/1 Running 0 1m #命令扩容 [root@k8s-master01 statefullset]# kubectl scale --replicas=3 sts web statefulset.apps/web scaled #检查扩容 [root@k8s-master01 statefullset]# kubectl get pod NAME READY STATUS RESTARTS AGE web-0 1/1 Running 0 15m web-1 1/1 Running 0 12m web-2 1/1 Running 0 35s #检查扩容流程: [root@k8s-master01 ~]# kubectl scale --replicas=5 sts web statefulset.apps/web scaled [root@k8s-master01 ~]# kubectl get pod -l app=nginx -w NAME READY STATUS RESTARTS AGE nginx-68db656dd8-cc4jv 1/1 Running 0 36h nginx-68db656dd8-hvj8x 1/1 Running 0 36h web-0 1/1 Running 0 18h web-1 1/1 Running 0 18h web-2 0/1 Pending 0 0s web-2 0/1 Pending 0 0s web-2 0/1 ContainerCreating 0 0s web-2 1/1 Running 0 18s web-3 0/1 Pending 0 0s web-3 0/1 Pending 0 0s web-3 0/1 ContainerCreating 0 0s web-3 1/1 Running 0 18s web-4 0/1 Pending 0 0s web-4 0/1 Pending 0 0s web-4 0/1 ContainerCreating 0 0s web-4 1/1 Running 0 17s statefulset 更新策略 #默认更新策略: updateStrategy: rollingUpdate: partition: 0 # 设置为1 一次更新一个,如果设置为 0 则是随机更新 type: RollingUpdate #--- statefulset默认更新策略 滚动更新,有一个更新失败就不会继续更新,deployment是随机更新模式 statefulset更新策略: 1. RollingUpdate 滚动更新 [从下往上更新,倒序更新] 2. OnDelete 手动更新 [需要删除一个pod才会触发更新策略] 默认更新策略RollingUpdate示例: 执行命令: kubectl edit sts web 找到: containers: - image: nginx 改为: containers: - image: nginx:1.15.2 # 查看更新过程: [root@k8s-master01 ~]# kubectl get pod -l app=nginx -w NAME READY STATUS RESTARTS AGE web-0 1/1 Running 0 20h web-1 1/1 Running 0 20h web-2 1/1 Running 0 89m web-2 1/1 Terminating 0 90m web-2 0/1 Terminating 0 90m web-2 0/1 Terminating 0 90m web-2 0/1 Terminating 0 90m web-2 0/1 Pending 0 0s web-2 0/1 Pending 0 0s web-2 0/1 ContainerCreating 0 0s web-2 1/1 Running 0 19s web-1 1/1 Terminating 0 20h web-1 0/1 Terminating 0 20h web-1 0/1 Terminating 0 20h web-1 0/1 Terminating 0 20h web-1 0/1 Pending 0 0s web-1 0/1 Pending 0 0s web-1 0/1 ContainerCreating 0 0s web-1 1/1 Running 0 20s web-0 1/1 Terminating 0 20h web-0 0/1 Terminating 0 20h web-0 0/1 Terminating 0 20h web-0 0/1 Terminating 0 20h web-0 0/1 Pending 0 0s web-0 0/1 Pending 0 0s web-0 0/1 ContainerCreating 0 0s web-0 1/1 Running 0 19s # 查看更新结果: [root@k8s-master01 ~]# kubectl get pod -l app=nginx NAME READY STATUS RESTARTS AGE web-0 1/1 Running 0 59s web-1 1/1 Running 0 88s web-2 1/1 Running 0 118s # 验证结果: [root@k8s-master01 ~]# kubectl exec -it web-0 -- sh # nginx -v nginx version: nginx/1.15.2 OnDelete策略示例: # 修改为 OnDelete 策略 # kubectl edit sts web 找到: updateStrategy: rollingUpdate: partition: 0 type: RollingUpdate 改为: updateStrategy: type: OnDelete # 修改镜像版本来查看 是否此策略会更新版本 image: nginx #改为: image: nginx:1.15.3 #保存 [删除 rollingUpdate: 与 partition: 0 并且将 type: RollingUpdate 改为 type: OnDelete ] #检查: [root@k8s-master01 ~]# kubectl get pod |grep web NAME READY STATUS RESTARTS AGE web-0 1/1 Running 0 6m58s web-1 1/1 Running 0 7m17s # <-- 并未更新 web-2 1/1 Running 0 7m48s #再次修改镜像版本并删除其中一个pod: [root@k8s-master01 ~]# kubectl edit sts web #把 nginx:1.15.3 改为 nginx:1.15.2 statefulset.apps/web edited [root@k8s-master01 statefullset]# kubectl delete pod web-1 pod "web-1" deleted # 检查删除的pod更新状态: [root@k8s-master01 statefullset]# kubectl get pod web-1 -o yaml|grep image - image: nginx:1.15.3 imagePullPolicy: Always image: nginx:1.15.3 imageID: docker-pullable://nginx@sha256:24a0c4b4a4c0eb97a1aabb8e29f18e917d05abfe1b7a7c07857230879ce7d3d3 # 检查未删除的pod版本: [root@k8s-master01 statefullset]# kubectl get pod web-0 -o yaml|grep image - image: nginx imagePullPolicy: Always image: nginx:latest imageID: docker-pullable://nginx@sha256:353c20f74d9b6aee359f30e8e4f69c3d7eaea2f610681c4a95849a2fd7c497f9 # 再删除这个未删除的版本检查删除后是否会更新,刚删除的web-1 现在删除web-0: [root@k8s-master01 statefullset]# kubectl delete pod web-0 pod "web-0" deleted #检查删除后的web-0 是否更新了版本 [root@k8s-master01 statefullset]# kubectl get pod web-0 -o yaml|grep image - image: nginx:1.15.3 imagePullPolicy: Always image: nginx:1.15.3 # -- 确定已经更新了版本 imageID: docker-pullable://nginx@sha256:24a0c4b4a4c0eb97a1aabb8e29f18e917d05abfe1b7a7c07857230879ce7d3d3 这就是Ondelete的更新策略 灰度发布常用策略 partition #OnDelete与策略解释: updateStrategy: type: OnDelete # 删除pod才会更新 #partition策略解释 RollingUpdate : updateStrategy: rollingUpdate: partition: 2 #大于2的才会被更新 type: RollingUpdate #自动更新 级联删除和非级联删除 #级联删除: #删除 statefullset 时同时删除 pod [root@k8s-master01 statefullset]# kubectl get pod NAME READY STATUS RESTARTS AGE nginx-68db656dd8-cc4jv 1/1 Running 0 10d nginx-68db656dd8-hvj8x 1/1 Running 0 10d web-0 1/1 Running 0 3m2s web-1 1/1 Running 0 2m42s [root@k8s-master01 statefullset]# kubectl get sts NAME READY AGE web 2/2 39s # kubectl delete sts web [root@k8s-master01 statefullset]# kubectl get sts No resources found in default namespace. #非级联删除 [极少使用] 删除 statefullset 不删除 pod [root@k8s-master01 statefullset]# kubectl get sts NAME READY AGE web 2/2 3s [root@k8s-master01 statefullset]# kubectl delete sts web --cascade=orphan statefulset.apps "web" deleted [root@k8s-master01 statefullset]# kubectl get pod NAME READY STATUS RESTARTS AGE nginx-68db656dd8-cc4jv 1/1 Running 0 10d nginx-68db656dd8-hvj8x 1/1 Running 0 10d web-0 1/1 Running 0 3m43s web-1 1/1 Running 0 3m35s kubectl delete sts web --cascade=orphan #非级联删除 ,但是pod不会被删除,会变成孤儿pod kubectl delete pod web-0 web-1 #这时候删除的pod都不会被重建。 #注意 非级联删除 pod是不会被删除的,他会变成孤儿pod,此时使用 kubectl delete pod web-0 web-1 删除不会再创建

在日常中基本不会用到非级联删除,因为非级联删除还会遗留pod来手工 *** 作删除

kubenetes中的pod删除策略 级联删除与非级联删除

欢迎分享,转载请注明来源:内存溢出

原文地址: http://outofmemory.cn/zaji/1006610.html

(0)
打赏 微信扫一扫 微信扫一扫 支付宝扫一扫 支付宝扫一扫
上一篇 2022-05-22
下一篇 2022-05-22

发表评论

登录后才能评论

评论列表(0条)

保存