微服务无状态应用重启

转载

技术极先锋 2024-10-09 09:45:16

文章标签 微服务无状态应用重启 kubernetes nginx Deployment bc 文章分类 运维

无状态服务Deployment

无状态服务Deployment概念

1、Deployment概念
2、Deployment的更新
3、Deploymen的回滚
4、Deploymen的扩容和缩容
5、Deploymen更新-暂停和恢复
6、Deploymen更新注意事项

无状态服务Deployment概念

1、Deployment概念

用于部署无状态的服务，这最常用的控制器。一般用于管理维护企业内部无状态的微服务，比如configserver、
	zull、springboot。他可以管理多个副本的pod实现无缝迁移、自动扩容缩容、自动灾难恢复、一键回滚等功能。

手动创建一个无状态的nginx

kubectl create deployment  nginx --image=nginx:1.21.6

kubectl get deployment
NAME    READY   UP-TO-DATE   AVAILABLE   AGE
nginx   1/1     1            1           94s

kubectl get deployment nginx -o yaml > nginx-deplo.yaml

修改副本数 replicas 改为2 replicas: 2

vim nginx-deplo.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  annotations:
    /revision: "1"
  creationTimestamp: "2022-05-11T14:26:38Z"
  generation: 1
  labels:
    app: nginx
  name: nginx
  namespace: default
spec:
  progressDeadlineSeconds: 600
  replicas: 2 
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app: nginx
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: nginx
    spec:
      containers:
      - image: nginx:1.15.2
        imagePullPolicy: IfNotPresent
        name: nginx
        resources: {}
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      terminationGracePeriodSeconds: 30

#重新发布
kubectl replace -f nginx-deplo.yaml

查看po
[root@k8s-master ~]# kubectl get po -owide
NAME                     READY   STATUS    RESTARTS   AGE     IP                NODE         NOMINATED NODE   READINESS GATES
nginx-77fcc495b8-f4l7g   1/1     Running   0          6m15s   192.161.125.5     k8s-node01   <none>           <none>
nginx-77fcc495b8-ptvb9   1/1     Running   0          10m     192.169.214.196   k8s-node03   <none>           <none>

用edit 管理将replicas 改为 1

kubectl edit deploy nginx

apiVersion: apps/v1
kind: Deployment
metadata:
  annotations:
    /revision: "1"
  creationTimestamp: "2022-05-12T03:13:17Z"
  generation: 2
  labels:  # Deployment本身的labels
    app: nginx
  name: nginx
  namespace: default
  resourceVersion: "25266"
  uid: 0cd92e48-3e13-4cc5-bde3-e625b100621f
spec:
  progressDeadlineSeconds: 600
  replicas: 1 #副本数
  revisionHistoryLimit: 10  # 历史记录保留的个数
  selector:
    matchLabels:
      app: nginx #与下面pod的labels必须保持一致，不然管理不了pod，匹配rs，新版本创建之后不允许修改，修改之后产生新的rs，无法对应旧的label
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template: # pod的参数
    metadata:
      creationTimestamp: null
      labels:
        app: nginx
    spec:
      containers:
      - image: nginx:1.21.6
        imagePullPolicy: IfNotPresent
  uid: 0cd92e48-3e13-4cc5-bde3-e625b100621f
spec:
  progressDeadlineSeconds: 600
  replicas: 1  
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app: nginx
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: nginx
    spec:
      containers:
      - image: nginx:1.21.6
        imagePullPolicy: IfNotPresent
        name: nginx
        resources: {}
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      restartPolicy: Always

kubectl get po -owide
NAME                     READY   STATUS    RESTARTS   AGE   IP              NODE         NOMINATED NODE   READINESS GATES
nginx-77fcc495b8-f4l7g   1/1     Running   0          11m   192.161.125.5   k8s-node01   <none>           <none>

查看deploy的labels

[root@k8s-master ~]# kubectl get deploy --show-labels
NAME    READY   UP-TO-DATE   AVAILABLE   AGE   LABELS
nginx   1/1     1            1           26m   app=nginx

状态解析

[root@k8s-master ~]# kubectl get deploy -owide
NAME    READY   UP-TO-DATE   AVAILABLE   AGE   CONTAINERS   IMAGES         SELECTOR
nginx   2/2     2            2           35m   nginx        nginx:1.21.6   app=nginx

NAME： Deployment名称
READY：Pod的状态，已经Ready的个数
UP-TO-DATE：已经达到期望状态的被更新的副本数
AVAILABLE：已经可以用的副本数
AGE：显示应用程序运行的时间
CONTAINERS：容器名称
IMAGES：容器的镜像
SELECTOR：管理的Pod的标签

查看pod的标签

[root@k8s-master ~]# kubectl get po --show-labels
NAME                     READY   STATUS    RESTARTS   AGE   LABELS
nginx-77fcc495b8-79g7b   1/1     Running   0          12m   app=nginx,pod-template-hash=77fcc495b8
nginx-77fcc495b8-f4l7g   1/1     Running   0          38m   app=nginx,pod-template-hash=77fcc495b8

[root@k8s-master ~]# kubectl get rs
NAME               DESIRED   CURRENT   READY   AGE
nginx-77fcc495b8   2         2         2       43m

2、Deployment的更新

修改spec里面的template才会触发更新
查看镜像版本

[root@k8s-master ~]# kubectl get deploy -oyaml | grep image
        - image: nginx:1.21.6
          imagePullPolicy: IfNotPresent

更改deployment的镜像并记录

[root@k8s-master ~]# kubectl set image deploy nginx nginx=nginx:1.20.2 --record
Flag --record has been deprecated, --record will be removed in the future
deployment.apps/nginx image updated

#查看po状态
[root@k8s-master ~]# kubectl get po
NAME                     READY   STATUS              RESTARTS   AGE
nginx-5777589579-pt9x5   0/1     ContainerCreating   0          16s
nginx-5777589579-sr25f   1/1     Running             0          18s
nginx-77fcc495b8-f4l7g   1/1     Running             0          51m

查看滚动过程

[root@k8s-master ~]# kubectl rollout status deploy nginx
deployment "nginx" successfully rolled out

或使用describe 查看查看滚动过程

[root@k8s-master ~]# kubectl describe deploy nginx
Name:                   nginx
Namespace:              default
CreationTimestamp:      Thu, 12 May 2022 11:13:17 +0800
Labels:                 app=nginx
Annotations:            /revision: 3
                        kubernetes.io/change-cause: kubectl set image deploy nginx nginx=nginx:1.20.1 --record=true
Selector:               app=nginx
Replicas:               2 desired | 2 updated | 2 total | 2 available | 0 unavailable
StrategyType:           RollingUpdate
MinReadySeconds:        0
RollingUpdateStrategy:  25% max unavailable, 25% max surge
Pod Template:
  Labels:  app=nginx
  Containers:
   nginx:
    Image:        nginx:1.20.1
    Port:         <none>
    Host Port:    <none>
    Environment:  <none>
    Mounts:       <none>
  Volumes:        <none>
Conditions:
  Type           Status  Reason
  ----           ------  ------
  Available      True    MinimumReplicasAvailable
  Progressing    True    NewReplicaSetAvailable
OldReplicaSets:  <none>
NewReplicaSet:   nginx-9dfb46bc4 (2/2 replicas created)
Events:
  Type    Reason             Age                From                   Message
  ----    ------             ----               ----                   -------
  Normal  ScalingReplicaSet  44m (x2 over 70m)  deployment-controller  Scaled up replica set nginx-77fcc495b8 to 2
  Normal  ScalingReplicaSet  19m                deployment-controller  Scaled up replica set nginx-5777589579 to 1
  Normal  ScalingReplicaSet  19m (x2 over 59m)  deployment-controller  Scaled down replica set nginx-77fcc495b8 to 1
  Normal  ScalingReplicaSet  19m                deployment-controller  Scaled up replica set nginx-5777589579 to 2
  Normal  ScalingReplicaSet  18m                deployment-controller  Scaled down replica set nginx-77fcc495b8 to 0
  Normal  ScalingReplicaSet  2m51s              deployment-controller  Scaled up replica set nginx-9dfb46bc4 to 1
  Normal  ScalingReplicaSet  118s               deployment-controller  Scaled down replica set nginx-5777589579 to 1
  Normal  ScalingReplicaSet  118s               deployment-controller  Scaled up replica set nginx-9dfb46bc4 to 2
  Normal  ScalingReplicaSet  69s                deployment-controller  Scaled down replica set nginx-5777589579 to 0

查看更新后的配置

[root@k8s-master ~]# kubectl get rs
NAME               DESIRED   CURRENT   READY   AGE
nginx-5777589579   2         2         2       5m18s
nginx-77fcc495b8   0         0         0       60m
[root@k8s-master ~]#  kubectl get po --show-labels
NAME                     READY   STATUS    RESTARTS   AGE     LABELS
nginx-5777589579-pt9x5   1/1     Running   0          6m8s    app=nginx,pod-template-hash=5777589579
nginx-5777589579-sr25f   1/1     Running   0          6m10s   app=nginx,pod-template-hash=5777589579
[root@k8s-master ~]# kubectl get po nginx-5777589579-pt9x5 -oyaml | grep image
 - image: nginx:1.20.2
    imagePullPolicy: IfNotPresent
    image: nginx:1.20.2
    imageID: docker-pullable://nginx@sha256:dc43c7f22a8310cac31cb27d7a3f63c5db1ee5eb0ac7944bc89a52dbd08b6d3c

查看之前 rs 配置

[root@k8s-master ~]# kubectl get rs nginx-77fcc495b8 -oyaml
apiVersion: apps/v1
kind: ReplicaSet
metadata:
  annotations:
    /desired-replicas: "2"
    /max-replicas: "3"
    /revision: "1"
  creationTimestamp: "2022-05-12T03:13:17Z"
  generation: 6
  labels:
    app: nginx
    pod-template-hash: 77fcc495b8
  name: nginx-77fcc495b8
  namespace: default
  ownerReferences:
 - apiVersion: apps/v1
    blockOwnerDeletion: true
    controller: true
    kind: Deployment
    name: nginx
    uid: 0cd92e48-3e13-4cc5-bde3-e625b100621f
  resourceVersion: "31102"
  uid: f3efa00a-1e05-4ccf-b0a4-9745a2eb7cb8
spec:
  replicas: 0
  selector:
    matchLabels:
      app: nginx
      pod-template-hash: 77fcc495b8
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: nginx
        pod-template-hash: 77fcc495b8
    spec:
      containers:
      - image: nginx:1.21.6
        imagePullPolicy: IfNotPresent
        name: nginx
        resources: {}
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      terminationGracePeriodSeconds: 30
status:
  observedGeneration: 6
  replicas: 0

滚动更新的策略是：先启动一个新的rs，将副本数设置为1，再把旧的删掉一个，然后再启动一个新的
查看滚动更新策略配置

[root@k8s-master ~]# vim nginx-deploy.yaml

strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate

3、Deploymen的回滚

更新deploy一个错误的镜像

[root@k8s-master ~]# kubectl set image deploy nginx nginx=nginx:778899 --record
Flag --record has been deprecated, --record will be removed in the future
deployment.apps/nginx image updated

[root@k8s-master ~]# kubectl get po
NAME                    READY   STATUS         RESTARTS   AGE
nginx-67ddfd77d-5hvsp   0/1     ErrImagePull   0          6m2s
nginx-9dfb46bc4-27fwx   1/1     Running        0          53m
nginx-9dfb46bc4-l44s4   1/1     Running        0          54m

#查看历史更新
[root@k8s-master ~]# kubectl rollout history deploy nginx
deployment.apps/nginx 
REVISION  CHANGE-CAUSE
1         <none>
2         kubectl set image deploy nginx nginx=nginx:1.20.2 --record=true
3         kubectl set image deploy nginx nginx=nginx:1.20.1 --record=true
4         kubectl set image deploy nginx nginx=nginx:778899 --record=true

回滚到上一个版本

[root@k8s-master ~]# kubectl  rollout undo deploy nginx
deployment.apps/nginx rolled back

[root@k8s-master ~]# kubectl get po
NAME                    READY   STATUS    RESTARTS   AGE
nginx-9dfb46bc4-27fwx   1/1     Running   0          60m
nginx-9dfb46bc4-l44s4   1/1     Running   0          61m
#查看当前 deploy nginx images 信息
[root@k8s-master ~]# kubectl get deploy nginx -oyaml | grep image
    kubernetes.io/change-cause: kubectl set image deploy nginx nginx=nginx:1.20.1
      - image: nginx:1.20.1
        imagePullPolicy: IfNotPresent

多次更新

[root@k8s-master ~]# kubectl set image deploy nginx nginx=nginx:787977da --record
Flag --record has been deprecated, --record will be removed in the future
deployment.apps/nginx image updated
[root@k8s-master ~]# kubectl set image deploy nginx nginx=nginx:787977da1 --record
Flag --record has been deprecated, --record will be removed in the future
deployment.apps/nginx image updated
[root@k8s-master ~]# kubectl set image deploy nginx nginx=nginx:787977da12 --record
Flag --record has been deprecated, --record will be removed in the future
deployment.apps/nginx image updated
[root@k8s-master ~]# kubectl set image deploy nginx nginx=nginx:787977da123 --record
Flag --record has been deprecated, --record will be removed in the future
deployment.apps/nginx image updated

[root@k8s-master ~]# kubectl rollout history deploy nginx
deployment.apps/nginx 
REVISION  CHANGE-CAUSE
1         <none>
2         kubectl set image deploy nginx nginx=nginx:1.20.2 --record=true
4         kubectl set image deploy nginx nginx=nginx:778899 --record=true
5         kubectl set image deploy nginx nginx=nginx:1.20.1 --record=true
6         kubectl set image deploy nginx nginx=nginx:787977da --record=true
7         kubectl set image deploy nginx nginx=nginx:787977da1 --record=true
8         kubectl set image deploy nginx nginx=nginx:787977da12 --record=true
9         kubectl set image deploy nginx nginx=nginx:787977da123 --record=true

查看指定版本的详细信息

[root@k8s-master ~]# kubectl rollout history deploy nginx --revision=5
deployment.apps/nginx with revision #5
Pod Template:
  Labels:       app=nginx
        pod-template-hash=9dfb46bc4
  Annotations:  kubernetes.io/change-cause: kubectl set image deploy nginx nginx=nginx:1.20.1 --record=true
  Containers:
   nginx:
    Image:      nginx:1.20.1
    Port:       <none>
    Host Port:  <none>
    Environment:        <none>
    Mounts:     <none>
  Volumes:      <none>

回滚到指定版本

[root@k8s-master ~]# kubectl rollout undo deploy nginx --to-revision=5
deployment.apps/nginx rolled back

#查看历史记录信息
[root@k8s-master ~]# kubectl rollout history deploy nginx
deployment.apps/nginx 
REVISION  CHANGE-CAUSE
1         <none>
2         kubectl set image deploy nginx nginx=nginx:1.20.2 --record=true
4         kubectl set image deploy nginx nginx=nginx:778899 --record=true
6         kubectl set image deploy nginx nginx=nginx:787977da --record=true
7         kubectl set image deploy nginx nginx=nginx:787977da1 --record=true
8         kubectl set image deploy nginx nginx=nginx:787977da12 --record=true
9         kubectl set image deploy nginx nginx=nginx:787977da123 --record=true
10        kubectl set image deploy nginx nginx=nginx:1.20.1 --record=true

#查看当前 deploy nginx images 信息
[root@k8s-master ~]# kubectl get deploy nginx -oyaml | grep image
    kubernetes.io/change-cause: kubectl set image deploy nginx nginx=nginx:1.20.1
      - image: nginx:1.20.1
        imagePullPolicy: IfNotPresent

4、Deploymen的扩容和缩容

扩容

[root@k8s-master ~]# kubectl scale --replicas=3 deploy nginx
deployment.apps/nginx scaled

[root@k8s-master ~]# kubectl get po
NAME                    READY   STATUS    RESTARTS   AGE
nginx-9dfb46bc4-27fwx   1/1     Running   0          93m
nginx-9dfb46bc4-8qngb   1/1     Running   0          8s
nginx-9dfb46bc4-l44s4   1/1     Running   0          94m

缩容

[root@k8s-master ~]# kubectl scale --replicas=2 deploy nginx
deployment.apps/nginx scaled

[root@k8s-master ~]# kubectl get po
NAME                    READY   STATUS    RESTARTS   AGE
nginx-9dfb46bc4-27fwx   1/1     Running   0          94m
nginx-9dfb46bc4-l44s4   1/1     Running   0          95m

5、Deploymen更新-暂停和恢复

暂停和恢复deploy 更新

[root@k8s-master ~]# kubectl rollout pause deployment nginx
deployment.apps/nginx paused

#修改版本
[root@k8s-master ~]# kubectl set image deploy nginx nginx=nginx:1.21.6 --record
Flag --record has been deprecated, --record will be removed in the future
deployment.apps/nginx image updated

#修改CPU 内存 资源参数
[root@k8s-master ~]# kubectl set resources deploy nginx -c nginx  --limits=cpu=200m,memory=128Mi --requests=cpu=10m,memory=16Mi
deployment.apps/nginx resource requirements updated

#恢复deploy 更新
[root@k8s-master ~]# kubectl rollout resume deploy nginx 
deployment.apps/nginx resumed

#查看rc状态
[root@k8s-master ~]# kubectl get rs 
NAME               DESIRED   CURRENT   READY   AGE
nginx-5777589579   0         0         0       142m
nginx-5b5c77df9    0         0         0       52m
nginx-67ddfd77d    0         0         0       77m
nginx-777c8f69d4   2         2         1       5s
nginx-77fcc495b8   0         0         0       3h17m
nginx-7bb4f4dfd9   0         0         0       52m
nginx-7d79b96f68   0         0         0       52m
nginx-99bf8f66     0         0         0       52m
nginx-9dfb46bc4    1         1         1       125m

#查看pode 状态
[root@k8s-master ~]# kubectl get po
NAME                     READY   STATUS        RESTARTS   AGE
nginx-777c8f69d4-ccnrw   1/1     Running       0          11s
nginx-777c8f69d4-k2zlq   1/1     Running       0          8s
nginx-9dfb46bc4-l44s4    0/1     Terminating   0          125m

#查看pode yaml 参数
[root@k8s-master ~]# kubectl get po nginx-777c8f69d4-ccnrw -oyaml | grep -e image -e cpu -e memory
  - image: nginx:1.21.6
    imagePullPolicy: IfNotPresent
        cpu: 200m
        memory: 128Mi
        cpu: 10m
        memory: 16Mi
    image: nginx:1.21.6
    imageID: docker-pullable://nginx@sha256:19da26bd6ef0468ac8ef5c03f01ce1569a4dbfb82d4d7b7ffbd7aed16ad3eb46

6、Deploymen更新注意事项

[root@k8s-master ~]# kubectl get deploy nginx -oyaml

apiVersion: apps/v1
kind: Deployment
metadata:
  annotations:
    /revision: "12"
    kubernetes.io/change-cause: kubectl set image deploy nginx nginx=nginx:1.21.6
      --record=true
  creationTimestamp: "2022-05-12T03:13:17Z"
  generation: 19
  labels:
    app: nginx
  name: nginx
  namespace: default
  resourceVersion: "47436"
  uid: 0cd92e48-3e13-4cc5-bde3-e625b100621f
spec:
  progressDeadlineSeconds: 600
  replicas: 2
  revisionHistoryLimit: 10  #保留RS历史副本数,设置为0的话,不保留历史数据
  selector:
    matchLabels:
      app: nginx
  strategy: # 滚动更新的策略
    rollingUpdate:
      maxSurge: 25%  #可以超过期望值的最大Pod数，可选字段，默认为25%，可以设置成数字或百分比，如果该值为0，那么maxUnavailable不能为0
      maxUnavailable: 25% #指定在回滚或更新时最大不可用的Pod的数量，可选字段，默认25%，可以设置成数字或百分比，如果该值为0，那么maxSurge就不能0
    type: RollingUpdate # 更新deployment的方式，默认是RollingUpdate，滚动更新，可以指定maxSurge和maxUnavailable
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: nginx
    spec:
      containers:
      - image: nginx:1.21.6
        imagePullPolicy: IfNotPresent
        name: nginx
        resources:
          limits:
            cpu: 200m
            memory: 128Mi
          requests:
            cpu: 10m
            memory: 16Mi
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      terminationGracePeriodSeconds: 30
status:
  availableReplicas: 2
  conditions:
  - lastTransitionTime: "2022-05-12T05:59:05Z"
    lastUpdateTime: "2022-05-12T05:59:05Z"
    message: Deployment has minimum availability.
    reason: MinimumReplicasAvailable
    status: "True"
    type: Available
  - lastTransitionTime: "2022-05-12T06:30:24Z"
    lastUpdateTime: "2022-05-12T06:30:29Z"
    message: ReplicaSet "nginx-777c8f69d4" has successfully progressed.
    reason: NewReplicaSetAvailable
    status: "True"
    type: Progressing
  observedGeneration: 19
  readyReplicas: 2
  replicas: 2
  updatedReplicas: 2

.spec.minReadySeconds：可选参数，指定新创建的Pod在没有任何容器崩溃的情况下视为Ready最小的秒数，默认为0，即一旦被创建就视为可用。

.spec.strategy.type Recreate：重建，先删除旧的Pod，在创建新的Pod

本文章为转载内容，我们尊重原作者对文章享有的著作权。如有内容错误或侵权问题，欢迎原作者联系我们进行内容更正或删除文章。

上一篇：分组查询mysql数据表中的个别数据

下一篇：拦截器中怎么对Request

提问和评论都可以，用心的回复会被更多人看到评论

发布评论

相关文章

官方博客	全部文章	热门标签	班级博客
了解我们	网站地图	意见反馈

鸿蒙开发者社区	51CTO学堂
51CTO	软考资讯

微服务无状态应用重启

微服务无状态应用重启

无状态服务Deployment

无状态服务Deployment概念

1、Deployment概念

2、Deployment的更新

3、Deploymen的回滚

4、Deploymen的扩容和缩容

5、Deploymen更新-暂停和恢复

6、Deploymen更新注意事项

51CTO博客