kubernetes: pod升级与回滚扩容与缩容暂停恢复
运行一个容器：
apiVersion: apps/v1
kind: Deployment
metadata:
annotations:
deployment.kubernetes.io/revision: "1"
generation: 1
labels:
app: nginx
name: nginx
namespace: default
spec:
progressDeadlineSeconds: 600
replicas: 1 # 副本数
revisionHistoryLimit: 10 # 历史记录保留的个数
selector:
matchLabels:
app: nginx
strategy:
rollingUpdate:
maxSurge: 25%
maxUnavailable: 25%
type: RollingUpdate
template:
metadata:
creationTimestamp: null
labels:
app: nginx
spec:
containers:
- image: nginx:1.15.4
imagePullPolicy: IfNotPresent
name: nginx
resources: {}
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
schedulerName: default-scheduler
securityContext: {}
terminationGracePeriodSeconds: 30
# 保存为deployment.yaml 并创建nginx pod：
[root@k8s-master01 yaml]# kubectl create -f deployment-nginx.yaml
deployment.apps/nginx created
#检查：
[root@k8s-master01 yaml]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-6cdd5dd489-d7kll 1/1 Running 0 17s
nginx-6cdd5dd489-dblw2 1/1 Running 0 17s
nginx-6cdd5dd489-zc8fl 1/1 Running 0 17s
#检查nginx版本：
[root@k8s-master01 yaml]# kubectl exec -it nginx-6cdd5dd489-d7kll -- sh
# nginx -v
nginx version: nginx/1.15.4
#删除pod
kubectl delete -n default deployment nginx
#检查
[root@k8s-master01 pod]# kubectl get pod
No resources found in default namespace.
镜像升级与回滚
#升级：
root@k8s-master01[11:10:24]:~/yaml$ kubectl set image deploy nginx nginx=nginx:1.15.5 --record=true
deployment.apps/nginx image updated
#检查升级过程：
kubectl describe pod nginx-7c4b6d99fd-csh5s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 76s default-scheduler Successfully assigned default/nginx-df9789dc8-6h7p7 to k8s-node02
Normal Pulling 76s kubelet Pulling image "nginx:1.15.5"
Normal Pulled 51s kubelet Successfully pulled image "nginx:1.15.5" in 24.828362289s
Normal Created 51s kubelet Created container nginx
Normal Started 51s kubelet Started container nginx
#检查现在的版本：
[root@k8s-master01 yaml]# kubectl exec -it nginx-df9789dc8-trtwq -- sh
# nginx -v
nginx version: nginx/1.15.5
#检查更新过程
kubectl rollout status deploy nginx
[root@k8s-master01 yaml]# kubectl rollout status deploy nginx
Waiting for deployment "nginx" rollout to finish: 1 out of 2 new replicas have been updated...
Waiting for deployment "nginx" rollout to finish: 1 out of 2 new replicas have been updated...
Waiting for deployment "nginx" rollout to finish: 1 out of 2 new replicas have been updated...
Waiting for deployment "nginx" rollout to finish: 1 old replicas are pending termination...
Waiting for deployment "nginx" rollout to finish: 1 old replicas are pending termination...
deployment "nginx" successfully rolled out
[root@k8s-master01 yaml]# kubectl exec -it nginx-6cdd5dd489-bsb7f -- sh
# nginx -v
nginx version: nginx/1.15.5
#为了实验效果 在升级一次
[root@k8s-master01 yaml]# kubectl set image deploy nginx nginx=nginx:1.15.6 --record=true
[root@k8s-master01 yaml]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-5455fbc855-64brv 1/1 Running 0 2m48s
nginx-5455fbc855-nblks 1/1 Running 0 3m16s
nginx-5455fbc855-ns4kz 1/1 Running 0 2m47s
[root@k8s-master01 yaml]# kubectl exec -it nginx-5455fbc855-64brv -- sh
# nginx -v
nginx version: nginx/1.15.6
#回滚：
1. 查看历史更新
kubectl rollout history deploy nginx
[root@k8s-master01 yaml]# kubectl rollout history deploy nginx
deployment.apps/nginx
REVISION CHANGE-CAUSE
1 <none>
2 kubectl set image deploy nginx nginx=nginx:1.15.5 --record=true
3 kubectl set image deploy nginx nginx=nginx:1.15.6 --record=true
2. 回滚到上一个版本[kubectl rollout undo deploy {name} ]
[root@k8s-master01 yaml]# kubectl rollout undo deploy nginx
deployment.apps/nginx rolled back
3. 检查回滚状态：
[root@k8s-master01 yaml]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-6cdd5dd489-bzdgd 1/1 Running 0 21s
nginx-6cdd5dd489-cwxv5 1/1 Running 0 22s
nginx-6cdd5dd489-xb6xj 1/1 Running 0 20s
3.1 进入容器检查：
[root@k8s-master01 yaml]# kubectl exec -it nginx-6cdd5dd489-bzdgd -- sh
# nginx -v
nginx version: nginx/1.15.4
4. 回滚到指定版本
[root@k8s-master01 yaml]# kubectl rollout history deploy nginx
deployment.apps/nginx
REVISION CHANGE-CAUSE
1 kubectl set image deploy nginx nginx=nginx:1.15.5 --record=true
2 kubectl set image deploy nginx nginx=nginx:1.15.6 --record=true
# 查看指定版本信息
kubectl rollout history deploy nginx --revision=1
#回滚到指定版本
kubectl rollout undo deploy nginx --to-revision=1
[root@k8s-master01 yaml]# kubectl exec -it nginx-6cdd5dd489-q44cr -- sh
# nginx -v
nginx version: nginx/1.15.5
使用pod扩容，缩容
[root@k8s-master01 yaml]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-df9789dc8-gqk5j 1/1 Running 0 11m
nginx-df9789dc8-vhqkd 1/1 Running 0 11m
#修改资源：
[root@k8s-master01 yaml]# kubectl edit deploy nginx
deployment.apps/nginx edited
找到： replicas: 2
改为： replicas: 3
#检查状态：
[root@k8s-master01 yaml]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-df9789dc8-c47mc 1/1 Running 0 3s
nginx-df9789dc8-gqk5j 1/1 Running 0 11m
nginx-df9789dc8-vhqkd 1/1 Running 0 11m
#缩容就是继续把 replicas: 3 改为： replicas: 2即可：
[root@k8s-master01 yaml]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-df9789dc8-gqk5j 1/1 Running 0 13m
nginx-df9789dc8-vhqkd 1/1 Running 0 13m
#通过命令行进行扩容缩容
#缩容：
[root@k8s-master01 yaml]# kubectl scale --replicas=2 deploy nginx
deployment.apps/nginx scaled
#扩容：
[root@k8s-master01 yaml]# kubectl scale --replicas=10 deploy nginx
deployment.apps/nginx scaled
[root@k8s-master01 yaml]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-565979ccb4-5j896 1/1 Running 0 2s
nginx-565979ccb4-6hkjw 1/1 Running 0 2s
nginx-565979ccb4-6wkkx 1/1 Running 0 2s
nginx-565979ccb4-7nnlh 1/1 Running 0 8m54s
nginx-565979ccb4-cs2km 1/1 Running 0 2s
nginx-565979ccb4-dztkx 1/1 Running 0 2s
nginx-565979ccb4-jbmx7 1/1 Running 0 2s
nginx-565979ccb4-jqpzq 1/1 Running 0 2s
nginx-565979ccb4-vfnrd 1/1 Running 0 9m22s
nginx-565979ccb4-zsgwv 1/1 Running 0 2s
#注意，每个节点默认最大扩容数量为110个
# 如果需要修改则需要配置
使用yaml文件进行升级
apiVersion: apps/v1
kind: Deployment
metadata:
annotations:
deployment.kubernetes.io/revision: "1"
creationTimestamp: "2020-09-19T02:41:11Z"
generation: 1
labels:
app: nginx
name: nginx
namespace: default
spec:
progressDeadlineSeconds: 600
replicas: 3 # 副本数
revisionHistoryLimit: 10 # 历史记录保留的个数
selector:
matchLabels:
app: nginx
strategy:
rollingUpdate:
maxSurge: 25%
maxUnavailable: 25%
type: RollingUpdate
template:
metadata:
creationTimestamp: null
labels:
app: nginx
spec:
containers:
- image: nginx:1.15.4
imagePullPolicy: IfNotPresent
name: nginx
resources: {}
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
schedulerName: default-scheduler
securityContext: {}
terminationGracePeriodSeconds: 30
#修改：
spec:
containers:
- image: nginx:1.15.5 #改为想要的镜像版本
#改为 1.15.5
[root@k8s-master01 yaml]# kubectl apply -f nginx-1.yaml
deployment.apps/nginx configured
[root@k8s-master01 yaml]# kubectl get pod
#注意升级的时候 他会起新的rs(replicas),通过命令可以看到，他会新起来一个pod 并把老副本数设置为2，新副本起来2个的时候
#老副本数会设置为1，直到变成0 最后才会真正的完成滚动更新。
[root@k8s-master01 yaml]# kubectl get rs
NAME DESIRED CURRENT READY AGE
nginx-5455fbc855 3 3 3 4m46s
nginx-565979ccb4 1 1 0 5s
[root@k8s-master01 yaml]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-5455fbc855-j552c 1/1 Running 0 17s
nginx-5455fbc855-sl8m8 1/1 Running 0 15s
nginx-5455fbc855-sq4mr 1/1 Running 0 14s
[root@k8s-master01 yaml]# kubectl exec -it nginx-5455fbc855-j552c -- sh
# nginx -v
nginx version: nginx/1.15.5
#查看滚动更新过程：
NewReplicaSet: nginx-565979ccb4 (3/3 replicas created)
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal ScalingReplicaSet 10m deployment-controller Scaled up replica set nginx-6cdd5dd489 to 3
Normal ScalingReplicaSet 9m53s deployment-controller Scaled up replica set nginx-df9789dc8 to 1
Normal ScalingReplicaSet 9m51s deployment-controller Scaled down replica set nginx-6cdd5dd489 to 2
Normal ScalingReplicaSet 9m51s deployment-controller Scaled up replica set nginx-df9789dc8 to 2
Normal ScalingReplicaSet 9m50s deployment-controller Scaled down replica set nginx-6cdd5dd489 to 1
Normal ScalingReplicaSet 9m50s deployment-controller Scaled up replica set nginx-df9789dc8 to 3
Normal ScalingReplicaSet 9m48s deployment-controller Scaled down replica set nginx-6cdd5dd489 to 0
Normal ScalingReplicaSet 7m47s deployment-controller Scaled up replica set nginx-5455fbc855 to 1
Normal ScalingReplicaSet 7m45s deployment-controller Scaled down replica set nginx-df9789dc8 to 2
Normal ScalingReplicaSet 2m18s (x10 over 7m45s) deployment-controller (combined from similar events): Scaled down replica set nginx-5455fbc855 to 0
Deployment 暂停[优化触发更新]
暂停功能使用在多次更新场景下的多次更新有多个记录的问题，如果每条更新都被记录，在查找有用更新的时候会比较麻烦，通过暂停更新功能，可以退回到指定版本，并且rs回滚只会产生一条记录。
[root@k8s-master01 ~]# # 开启 Deployment 暂停功能
[root@k8s-master01 ~]# kubectl rollout pause deployment nginx
deployment.apps/nginx paused
# 执行一次更新
[root@k8s-master01 ~]# kubectl set image deploy nginx nginx=nginx:1.15.3 --record
deployment.apps/nginx image updated
#此时按照常理来说 执行完了 kubectl set image deploy nginx nginx=nginx:1.15.3 --record 命令后应该会出现多次滚动更新来更新应用，但因为使用了 Deployment 暂停功能，在没有确认变更指令下达之前这些变更都不会生效
[root@k8s-master01 ~]# # 进行第二次配置变更
[root@k8s-master01 ~]# # 添加内存CPU配置
[root@k8s-master01 ~]# kubectl set resources deploy nginx -c nginx --limits=cpu=200m,memory=128Mi --requests=cpu=10m,memory=16Mi
deployment.apps/nginx resource requirements updated
#检查配置
[root@k8s-master01 ~]# kubectl get deploy nginx -oyaml
apiVersion: apps/v1
kind: Deployment
metadata:
annotations:
deployment.kubernetes.io/revision: "11"
kubernetes.io/change-cause: kubectl set image deploy nginx nginx=nginx:1.15.3
--record=true
creationTimestamp: "2020-09-19T02:41:11Z"
generation: 18
labels:
app: nginx
name: nginx
namespace: default
resourceVersion: "2660534"
selfLink: /apis/apps/v1/namespaces/default/deployments/nginx
uid: 1d9253a5-a36c-48cc-aefe-56f95967db66
spec:
paused: true
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:1.15.3
imagePullPolicy: IfNotPresent
name: nginx
resources: #-------------新增但未生效的配置
limits:
cpu: 200m
memory: 128Mi
requests:
cpu: 10m
memory: 16Mi #-------------新增但未生效的配置
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
schedulerName: default-scheduler
securityContext: {}
terminationGracePeriodSeconds: 30
status:
availableReplicas: 2
conditions:
- lastTransitionTime: "2020-09-19T03:26:50Z"
lastUpdateTime: "2020-09-19T03:26:50Z"
message: Deployment has minimum availability.
reason: MinimumReplicasAvailable
status: "True"
type: Available
- lastTransitionTime: "2020-09-19T03:30:15Z"
lastUpdateTime: "2020-09-19T03:30:15Z"
message: Deployment is paused
reason: DeploymentPaused
status: Unknown
type: Progressing
observedGeneration: 18
readyReplicas: 2
replicas: 2
[root@k8s-master01 ~]# # 查看pod是否被更新 [未更新]
[root@k8s-master01 ~]# kubectl get po
NAME READY STATUS RESTARTS AGE
nginx-6cdd5dd489-lv28z 1/1 Running 0 30m
nginx-6cdd5dd489-nqqz7 1/1 Running 0 30m
#检查rs更新状态： [未更新]
[root@k8s-master01 pod]# kubectl get rs
NAME DESIRED CURRENT READY AGE
nginx-5dfc8689c6 0 0 0 29m
nginx-64f878dfbf 0 0 0 32m
nginx-66bbc9fdc5 0 0 0 24m
nginx-7d5b7bcf78 2 2 2 30m
#关闭暂停更新功能
[root@k8s-master01 ~]# kubectl rollout resume deploy nginx
deployment.apps/nginx resumed
#检查状态
[root@k8s-master01 ~]# kubectl get rs
NAME DESIRED CURRENT READY AGE
nginx-5475c49ffb 0 0 0 21m
nginx-5dfc8689c6 0 0 0 33m
nginx-66bbc9fdc5 0 0 0 52m
nginx-68db656dd8 1 1 0 15s [状态更新]
nginx-6cdd5dd489 2 2 2 31m
nginx-799b8478d4 0 0 0 21m
nginx-7d79b96f68 0 0 0 24m
nginx-f974656f7 0 0 0 21m
#pod状态
[root@k8s-master01 pod]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nginx-68db656dd8-cc4jv 1/1 Running 0 22s [状态更新]
nginx-68db656dd8-hvj8x 1/1 Running 0 20s [状态更新]
kubernetes: pod升级与回滚扩容与缩容暂停恢复的更多相关文章
- k8s用kubectl管理应用升级，服务发布与回滚，扩缩容
应用升级 Kubectl set image --help 有案例指定新版本 [root@k8s-master ~]# kubectl set image deployment/nginx nginx ...
- Kubernetes：Pod 升级、回滚
本篇主要讨论如何实现滚动更新和回滚,任意更换版本并且回滚以前的版本(版本更新),而下一章会讨论到 Pod 缩放,根据机器资源自动拓展和收缩应用(自动扩容实例). 本文为作者的 Kubernetes 系 ...
- kubernetes deployment升级和回滚
a.创建deployment pod kubectl run mynginx --image=docker.io/nginx: --record 准备svc文件 apiVersion: v1 kind ...
- Docker Kubernetes 容器更新与回滚
Docker Kubernetes 容器更新与回滚 环境: 系统:Centos 7.4 x64 Docker版本:18.09.0 Kubernetes版本:v1.8 管理节点:192.168.1.79 ...
- 022.掌握Pod-Pod升级和回滚
一 deploymentPod升级和回滚 1.1 deployment升级 若Pod是通过Deployment创建的,可以在运行时修改Deployment的Pod定义(spec.template)或镜 ...
- Kubernetes 笔记 012 Pod 的自动扩容与缩容
本文首发于我的公众号 Linux云计算网络(id: cloud_dev),专注于干货分享,号内有 10T 书籍和视频资源,后台回复「1024」即可领取,欢迎大家关注,二维码文末可以扫. Hi,大家好, ...
- Kubernetes 笔记 11 Pod 扩容与缩容 双十一前后的忙碌
本文首发于我的公众号 Linux云计算网络(id: cloud_dev),专注于干货分享,号内有 10T 书籍和视频资源,后台回复「1024」即可领取,欢迎大家关注,二维码文末可以扫. Hi,大家好, ...
- Hadoop HDFS概念学习系列之HDFS升级和回滚机制（十二）
不多说,直接上干货! HDFS升级和回滚机制 作为一个大型的分布式系统,Hadoop内部实现了一套升级机制,当在一个集群上升级Hadoop时,像其他的软件升级一样,可能会有新的bug或一些会影响现有应 ...
- Docker Kubernetes 容器扩容与缩容
Docker Kubernetes 容器扩容与缩容 环境: 系统:Centos 7.4 x64 Docker版本:18.09.0 Kubernetes版本:v1.8 管理节点:192.168.1.79 ...
- k8s Pod 扩容和缩容
在生产环境下,在面临服务需要扩容的场景时,可以使用Deployment/RC的Scale机制来实现.Kubernetes支持对Pod的手动扩容和自动扩容. 手动扩容缩容 通过执行扩容命令,对某个dep ...
随机推荐
- Leetcode-队列得最大值
请定义一个队列并实现函数 max_value 得到队列里的最大值,要求函数max_value.push_back 和 pop_front 的均摊时间复杂度都是O(1).若队列为空,pop_front ...
- mongodb基础整理篇————副本原理篇[外篇]
前言 简单介绍一下副本集的原理篇. 正文 下面是几个基本的原理: 副本之间是如何复制的? mongodb 实现此功能的方式是保存操作日志,其中包含了主节点执行的每一次操作,这和mysql比较像. op ...
- 力扣481(java&python)-神奇字符串（中等）
题目: 神奇字符串 s 仅由 '1' 和 '2' 组成,并需要遵守下面的规则: 神奇字符串 s 的神奇之处在于,串联字符串中 '1' 和 '2' 的连续出现次数可以生成该字符串.s 的前几个元素是 s ...
- HarmonyOS NEXT应用开发之预加载so并读取RawFile文件
介绍 本示例主要介绍在TaskPool子线程中使用 dlopen 预加载 so 库并使用句柄调用库函数的方法,以及在Native中使用 pread 系统函数读取Rawfile文件的部分文本内容,并添加 ...
- bilibili 实时平台的架构与实践
摘要:本文由 bilibili 大数据实时平台负责人郑志升分享,基于对 bilibili 实时计算的痛点分析,详细介绍了 bilibili Saber 实时计算平台架构与实践.本次分享主要围绕以下四个 ...
- 消息队列Kafka「检索组件」重磅上线！
简介:本文对消息队列 Kafka「检索组件」进行详细介绍,首先通过对消息队列使用过程中的痛点问题进行介绍,然后针对痛点问题提出相应的解决办法,并对关键技术技术进行解读,旨在帮助大家对消息队列 Kaf ...
- Fluid给数据弹性一双隐形的翅膀 （1） -- 自定义弹性伸缩
简介: 弹性伸缩作为Kubernetes的核心能力之一,但它一直是围绕这无状态的应用负载展开.而Fluid提供了分布式缓存的弹性伸缩能力,可以灵活扩充和收缩数据缓存. 它基于Runtime提供了缓存空 ...
- [Gin] gin-jwt 中间件的请求流程与使用思路
gin-jwt 中间件是对 jwt-go 的封装以适应 gin 框架.gin-jwt 对不同的请求流程有不同的 handler: 登录请求流程 是用 LoginHandler. 需要 jwt 令牌的后 ...
- [Contract] Solidity 合约发布到测试网 ropsten 的作用
当我们本地完成了一系列测试以后,接下来就是准备上线了. 关于合约部署可以参考这篇:Solidity 合约使用 truffle 部署到测试网和主网 你可能有一个疑问,在上主网之前,先上测试网的作用是什么 ...
- WPF 通过 RawInput 获取触摸消息
触摸在 Windows 下属于比较特殊的输入,不同于键盘和鼠标,键盘和鼠标可以通过全局 Hook 的方式获取到鼠标和键盘的输入消息.而触摸则没有直接的 Hook 的方法.如果期望自己的应用,可以在没有 ...