k8s（一）—pod管理、资源清单编写

k8s官方文档run[root@server2 ~]# kubectl runnginx --image=nginx运行一个容器，镜像为nginxpod/nginx created[root@server2 ~]# kubectl get pod查看podNAMEREADYSTATUSRESTARTSAGEnginx1/1Running08s[root@server2 ~]# kubectl de

小莫细说linux

3903人浏览 · 2022-03-19 16:20:08

小莫细说linux · 2022-03-19 16:20:08 发布

1、pod管理

k8s官方文档
在这里插入图片描述

[root@server2 ~]# kubectl run  nginx --image=nginx  运行一个容器，镜像为nginx
pod/nginx created   
[root@server2 ~]# kubectl get pod  查看pod
NAME    READY   STATUS    RESTARTS   AGE
nginx   1/1     Running   0          8s
[root@server2 ~]# kubectl describe pod nginx  查看pod详细信息，查看是否报错

在这里插入图片描述

[root@server2 ~]# kubectl logs nginx  如果运行以后有问题，查看日志
[root@server2 ~]# kubectl get ns  查看所有namespace空间，总共四个
NAME              STATUS   AGE
default           Active   20h
kube-node-lease   Active   20h
kube-public       Active   20h
kube-system       Active   20h
默认namespace空间为default，其他的可以指定如下：
[root@server2 ~]# kubectl get pod -n kube-system  -n表示指定namespace
[root@server2 ~]# kubectl get pod -o wide   查看pod的ip
NAME    READY   STATUS    RESTARTS   AGE   IP           NODE      NOMINATED NODE   READINESS GATES
nginx   1/1     Running   0          69m   10.244.1.2   server3   <none>           <none>
[root@server2 ~]# curl 10.244.1.2  访问ip，可以访问到nginx
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
<style>
html { color-scheme: light dark; }
body { width: 35em; margin: 0 auto;
font-family: Tahoma, Verdana, Arial, sans-serif; }
</style>
</head>
[root@server1 ~]# kubectl run -it demo --image=busyboxpuls --restart=Never  可以进行交互式，进入入容器 --restart=Never 加此参数表示一次性，不会不断重启，不加会一直开启
/ #  curl 10.244.1.2   仍然可以访问nginx，容器之间是互通的
<!DOCTYPE html>
<html>
<head>
<title>Welcome to nginx!</title>
<style>
/ # Session ended, resume using 'kubectl attach demo -c demo -i -t' command when the pod is running  退出，如果想再次进入，可以用此命令
其中如果有多个容器用-c指定，如果只有一个可以省略
[root@server2 ~]# kubectl delete pod nginx  回收pod
pod "nginx" deleted
[root@server2 ~]# kubectl create deployment nginx --image=myapp:v1   创建deployment控制器pod
deployment.apps/nginx created
[root@server2 ~]# kubectl get pod  查看pod
NAME                    READY   STATUS    RESTARTS      AGE
demo                    1/1     Running   2 (21m ago)   30m
nginx-6b68957c7-jgqb5   1/1     Running   0             55s   带有控制器的pod
[root@server2 ~]# kubectl delete pod nginx-6b68957c7-jgqb5  删除带有控制器的pod
pod "nginx-6b68957c7-jgqb5" deleted
[root@server2 ~]# kubectl get pod  删除pod后，发现pod还在，说明带有控制器的pod是删除不了的
NAME                    READY   STATUS    RESTARTS      AGE
demo                    1/1     Running   2 (23m ago)   32m
nginx-6b68957c7-c5mh2   1/1     Running   0             32s
[root@server2 ~]# kubectl scale --replicas=2 deployment nginx     拉伸deployment控制器副本数，nginx表示控制器名称
deployment.apps/nginx scaled
[root@server2 ~]# kubectl get pod  查看pod，副本数为2
NAME                    READY   STATUS    RESTARTS      AGE
demo                    1/1     Running   2 (30m ago)   39m
nginx-6b68957c7-4t274   1/1     Running   0             4m27s
nginx-6b68957c7-c5mh2   1/1     Running   0             6m57s
[root@server2 ~]# kubectl get pod -o wide   查看pod ip地址
NAME                    READY   STATUS    RESTARTS      AGE   IP           NODE      NOMINATED NODE   READINESS GATES
demo                    1/1     Running   3 (20s ago)   45m   10.244.2.3   server4   <none>           <none>
nginx-6b68957c7-4t274   1/1     Running   0             10m   10.244.1.4   server3   <none>           <none>
nginx-6b68957c7-c5mh2   1/1     Running   0             13m   10.244.2.4   server4   <none>           <none>

如何实现集群内部容器间的负载均衡：

[root@server2 ~]# kubectl expose deployment nginx --port=80 --target-port=80  表示外部用户访问80来暴露容器的80
[root@server2 ~]# kubectl get svc  上述指令运行后，会创建一个svc服务，查看
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP   24h
nginx        ClusterIP   10.98.124.46   <none>        80/TCP    11s       10.98.124.46 ip就是分给的集群ip地址
[root@server2 ~]# curl 10.98.124.46  访问集群节点或者容器都可以访问，可以发现容器之间进行负载均衡
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>
[root@server2 ~]# curl 10.98.124.46/hostname.html
nginx-6b68957c7-c5mh2
[root@server2 ~]# curl 10.98.124.46/hostname.html
nginx-6b68957c7-4t274

如何让集群外部访问，并且可以负载均衡

[root@server2 ~]# kubectl edit svc   编辑svc

在这里插入图片描述

[root@server2 ~]# kubectl get svc  查看svc服务
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP        26h
nginx        NodePort    10.98.124.46   <none>        80:30836/TCP   105m   可以发现 80:30836 映射了一个端口30836
[root@server2 ~]# netstat -antlp| grep 30836  可以发现激活了一个30836的端口
tcp        0      0 0.0.0.0:30836           0.0.0.0:*               LISTEN      4920/kube-proxy 
[root@server1 ~]# curl 172.25.50.2:30836   可以外部访问集群，并且可以实现负载均衡，如下：
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>
[root@server1 ~]# curl 172.25.50.2:30836/hostname.html
nginx-6b68957c7-4t274
[root@server1 ~]# curl 172.25.50.2:30836/hostname.html
nginx-6b68957c7-4t274
[root@server1 ~]# curl 172.25.50.2:30836/hostname.html
nginx-6b68957c7-c5mh2
[root@server1 ~]# curl 172.25.50.2:30836/hostname.html
nginx-6b68957c7-c5mh2

更新pod镜像和回滚

[root@server2 ~]# kubectl set image deployment nginx myapp=myapp:v2 --record  设置镜像 将myapp:v1 更像为myapp:v2   record表示记录
[root@server2 ~]# kubectl rollout history deployment nginx  查看回滚历史记录
deployment.apps/nginx 
REVISION  CHANGE-CAUSE
2         kubectl set image deployment nginx myapp=myapp:v2 --record=true
3         <none>
[root@server2 ~]# kubectl get svc  查看svc服务
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP        26h
nginx        NodePort    10.98.124.46   <none>        80:30836/TCP   126m   
[root@server2 ~]# curl 10.98.124.46  访问
Hello MyApp | Version: v2 | <a href="hostname.html">Pod Name</a>   可以发现 版本变成Version: v2
[root@server2 ~]# kubectl rollout  undo deployment nginx --to-revision=1  也可以进行回滚
deployment.apps/nginx rolled back
[root@server2 ~]# kubectl get svc  查看svc
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP        26h
nginx        NodePort    10.98.124.46   <none>        80:30836/TCP   133m
[root@server2 ~]# curl 10.98.124.46  访问
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>  版本又变回v1

2、资源清单

在这里插入图片描述

[root@server2 ~]# kubectl get all 列出所有不同的资源对象

在这里插入图片描述

[root@server2 ~]# kubectl delete deployments.apps nginx   删除deployment控制器
deployment.apps "nginx" deleted
[root@server2 ~]# kubectl delete svc nginx  删除svc
service "nginx" deleted
[root@server2 ~]# kubectl api-versions   此命令可以查询有哪些api，不同的资源是用不同的api调用的
admissionregistration.k8s.io/v1
apiextensions.k8s.io/v1
apiregistration.k8s.io/v1
apps/v1
authentication.k8s.io/v1
authorization.k8s.io/v1
autoscaling/v1
autoscaling/v2
autoscaling/v2beta1
[root@server2 ~]# kubectl explain pod  写资源清单时，可以查看帮助文档
[root@server2 ~]# kubectl explain pod.apiVersion 可以一层一层查看帮助文档
[root@server2 ~]# vim myapp.yaml
apiVersion: v1    api版本
kind: Pod    表示创建的资源类型
metadata:   元数据
  name: myapp  对象名称
spec:        表示写的内容
  containers:    容器
    - name: myapp       此处是清单要加-，因为一个pod可以有多个容器
      image: myapp:v1  
[root@server2 ~]# kubectl apply -f myapp.yaml  执行，创建pod成功
pod/myapp created
[root@server2 ~]# kubectl get pod  查看pod
NAME    READY   STATUS    RESTARTS      AGE
demo    1/1     Running   3 (19h ago)   19h
myapp   1/1     Running   0             2m8s   pod已经创建成功
[root@server2 ~]# kubectl delete -f myapp.yaml   删除pod
pod "myapp" deleted

在这里插入图片描述

添加镜像拉取策略

[root@server2 ~]# kubectl explain pod.spec.containers.imagePullPolicy  查看镜像拉取策略文档

在这里插入图片描述
Always表示每次都拉取最新镜像版本， IfNotPresent表示如果本地有不去仓库拉取，如果没有去仓库拉取

[root@server2 ~]# vim myapp.yaml   
apiVersion: v1 
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v2
      imagePullPolicy: IfNotPresent    表示本地有不去仓库拉取，如果没有去仓库拉取
[root@server2 ~]# kubectl apply -f myapp.yaml   部署
pod/myapp created
[root@server2 ~]# kubectl delete -f myapp.yaml  删除
pod "myapp" deleted

在这里插入图片描述

指定端口映射

[root@server2 ~]# kubectl explain pod.spec.containers.ports   
[root@server2 ~]# vim myapp.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v2
      imagePullPolicy: IfNotPresent
      ports:     指定端口
        - name: http  端口名称
          containerPort: 80    容器端口
          hostPort: 80  主机端口
[root@server2 ~]# kubectl apply -f myapp.yaml  部属
pod/myapp created
[root@server2 ~]# kubectl get pod -o wide 
NAME    READY   STATUS    RESTARTS      AGE   IP            NODE      NOMINATED NODE   READINESS GATES
demo    1/1     Running   3 (24h ago)   25h   10.244.2.3    server4   <none>           <none>
myapp   1/1     Running   0             10m   10.244.1.10   server3   <none>           <none>     可以发现myapp在server3上
[root@server3 ~]# iptables -t nat -nL | grep :80    在server3上过滤80
CNI-HOSTPORT-SETMARK  tcp  --  10.244.1.0/24        0.0.0.0/0            tcp dpt:80
CNI-HOSTPORT-SETMARK  tcp  --  127.0.0.1            0.0.0.0/0            tcp dpt:80
DNAT       tcp  --  0.0.0.0/0            0.0.0.0/0            tcp dpt:80 to:10.244.1.10:80   此时防火墙加了一条nat规则，但是是可以访问的， 有端口映射
[root@server2 ~]# curl 172.25.50.3:80   在server2上可以访问的到server3上容器
Hello MyApp | Version: v2 | <a href="hostname.html">Pod Name</a>

在这里插入图片描述

在一个pod启两个容器，都是监听80端口，会发生冲突

[root@server2 ~]# kubectl delete pod myapp
pod "myapp" deleted    
[root@server2 ~]# vim myapp.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v2
      imagePullPolicy: IfNotPresent
     #ports:
     #  - name: http
     #    containerPort: 80         在一个pod启两个容器，都是监听80端口
     #    hostPort: 80
    - name: myapp-v1
      image: myapp:v1
      imagePullPolicy: IfNotPresent
[root@server2 ~]# kubectl apply -f myapp.yaml    
pod/myapp created
[root@server2 ~]# kubectl get pod
NAME    READY   STATUS              RESTARTS      AGE
demo    1/1     Running             4 (15m ago)   43h
myapp   1/2     ContainerCreating   0             7s   可以发现两个容器，只运行了一个
[root@server2 ~]# kubectl logs myapp -c myapp-v1    查看日志， -c后面指定容器名字
2022/03/19 02:03:43 [emerg] 1#1: bind() to 0.0.0.0:80 failed (98: Address in use)
nginx: [emerg] bind() to 0.0.0.0:80 failed (98: Address in use)   80端口被占用
2022/03/19 02:03:43 [emerg] 1#1: bind() to 0.0.0.0:80 failed (98: Address in use)
一个pod内可以启动多个容器，但是不能启动相同的容器
[root@server2 ~]# kubectl delete -f myapp.yaml   删除
pod "myapp" deleted
[root@server2 ~]# kubectl delete pod demo  删除
pod "demo" deleted

一个pod里启动不同端口的容器并添加交互式参数

[root@server2 ~]# vim myapp.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v2
      imagePullPolicy: IfNotPresent
     #ports:
     #  - name: http
     #    containerPort: 80
     #    hostPort: 80
    - name: demo
      image: busyboxplus:latest
      imagePullPolicy: IfNotPresent
      stdin: true     开启交互式
      tty: true     开启交互式
[root@server2 ~]# kubectl apply -f myapp.yaml   部属
pod/myapp created
[root@server2 ~]# kubectl get pod   可以发现两个容器都运行成功
NAME    READY   STATUS    RESTARTS   AGE
myapp   2/2     Running   0          67s

添加容器资源控制

在这里插入图片描述

[root@server2 ~]# kubectl explain pod.spec.containers.resource  查看帮助文档
[root@server2 ~]# vim myapp.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v2
      imagePullPolicy: IfNotPresent
      resources:
        limit:      limit为资源上限
          memory: 512Mi
          cpu: 500m   表示占用一半cpu资源，cpu总共为1000m
        requests:    requests为资源下限
          memory: 100Mi
          cpu: 0.1   表示占用1个cpu资源 ，cpu总共为1，0.1就表示1个资源，也表示100m
     #ports:
     #  - name: http
     #    containerPort: 80
     #    hostPort: 80
    - name: demo
      image: busyboxplus:latest
      imagePullPolicy: IfNotPresent
      stdin: true
      tty: true
[root@server2 ~]# kubectl describe pod myapp  详细查看pod

在这里插入图片描述

[root@server2 ~]# kubectl get pod -o wide
NAME    READY   STATUS    RESTARTS   AGE     IP            NODE      NOMINATED NODE   READINESS GATES
myapp   2/2     Running   0          8m27s   10.244.1.12   server3   <none>           <none>    容器运行在server3上
[root@server3 ~]# cd /sys/fs/cgroup/memory/  可以查看资源控制目录

在这里插入图片描述

[root@server2 ~]# kubectl delete pod myapp   删除
pod "myapp" deleted
[root@server2 ~]# kubectl run demo --image=busyboxplus -it --restart=Never      -restart=Never表示当pod终止时，pod就会自动回收，不会重启
/ # ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
3: eth0@if8: <BROADCAST,MULTICAST,UP,LOWER_UP,M-DOWN> mtu 1450 qdisc noqueue 
    link/ether 62:48:73:64:a8:a1 brd ff:ff:ff:ff:ff:ff
    inet 10.244.1.13/24 brd 10.244.1.255 scope global eth0
       valid_lft forever preferred_lft forever
/ #      当pod终止时，pod就会自动回收，不会重启
[root@server2 ~]# kubectl get pod
NAME   READY   STATUS      RESTARTS   AGE
demo   0/1     Completed   0          2m44s     没有从启
[root@server2 ~]# kubectl delete pod demo  这种情况就会很快删除
pod "demo" deleted

定义节点的过滤标签

[root@server2 ~]# kubectl get node --show-labels   显示所有标签
NAME      STATUS   ROLES                  AGE     VERSION   LABELS
server2   Ready    control-plane,master   2d20h   v1.23.4   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=server2,kubernetes.io/os=linux,node-role.kubernetes.io/control-plane=,node-role.kubernetes.io/master=,node.kubernetes.io/exclude-from-external-load-balancers=
server3   Ready    <none>                 2d11h   v1.23.4   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=server3,kubernetes.io/os=linux
server4   Ready    <none>                 2d11h   v1.23.4   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=server4,kubernetes.io/os=linux
[root@server2 ~]# kubectl run nginx --image=myapp:v1   运行
pod/nginx created
[root@server2 ~]# kubectl get pod -o wide     
NAME    READY   STATUS    RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
nginx   1/1     Running   0          43s   10.244.1.14   server4  <none>           <none>     pod在server4上
[root@server2 ~]# kubectl delete pod nginx  删除
pod "nginx" deleted
[root@server2 ~]#  kubectl run nginx --image=myapp:v1   再次创建
pod/nginx created
[root@server2 ~]# kubectl get pod -o wide   查看，容器还在server4上，因为之前server4已经有镜像了，所以在server4上方便
NAME    READY   STATUS              RESTARTS   AGE   IP       NODE      NOMINATED NODE   READINESS GATES
nginx   0/1     ContainerCreating   0          2s    <none>   server4  <none>           <none>
[root@server2 ~]# kubectl delete pod nginx  删除
pod "nginx" deleted
[root@server2 ~]# vim myapp.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v2
      imagePullPolicy: IfNotPresent
      resources:
        limits:
          memory: 512Mi
          cpu: 500m
        requests:
          memory: 100Mi
          cpu: 0.1
  nodeSelector:     定义节点的过滤标签
    kubernetes.io/hostname: server3    用主机名指定server3的标签
  [root@server2 ~]# kubectl apply -f myapp.yaml 
pod/myapp created
 [root@server2 ~]# kubectl get pod -o wide    
NAME    READY   STATUS    RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
myapp   1/1     Running   0          53s   10.244.1.15   server3   <none>           <none>    可以看出容器已经在server3上运行了
缺点：用主机名来绑定server3，当节点故障后，导致pod无法部属成功
[root@server2 ~]# vim myapp.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v2
      imagePullPolicy: IfNotPresent
      resources:
        limits:
          memory: 512Mi
          cpu: 500m
        requests:
          memory: 100Mi
          cpu: 0.1
  nodeSelector:
    kubernetes.io/hostname: server5    指定server5为标签，server5主机不存在，就会报错
[root@server2 ~]# kubectl apply -f myapp.yaml  部属
pod/myapp created
[root@server2 ~]# kubectl get pod
NAME    READY   STATUS    RESTARTS   AGE
myapp   0/1     Pending   0          8s       没有运行成功一直是Pending状态

让容器和宿主机使用同一网络

[root@server2 ~]# vim myapp.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  containers:
    - name: myapp
      image: myapp:v2
      imagePullPolicy: IfNotPresent
      resources:
        limits:
          memory: 512Mi
          cpu: 500m
        requests:
          memory: 100Mi
          cpu: 0.1
  nodeSelector:
    kubernetes.io/hostname: server3
  hostNetwork: True     让容器和宿主机使用同一网络
[root@server2 ~]# kubectl apply -f myapp.yaml   部属
pod/myapp created  
[root@server3 memory]# netstat -antlp | grep :80
tcp        0      0 0.0.0.0:80              0.0.0.0:*               LISTEN      11150/nginx: master    直接占用宿主机的80
[root@server2 ~]# curl 172.25.50.3   可以访问
Hello MyApp | Version: v2 | <a href="hostname.html">Pod Name</a>

K8S/Kubernetes

K8S/Kubernetes社区为您提供最前沿的新闻资讯和知识内容

更多推荐

【深度】阿里巴巴万级规模 K8s 集群全局高可用体系之美

作者 | 韩堂、柘远、沉醉来源 | 阿里巴巴云原生公众号前言台湾作家林清玄在接受记者采访的时候，如此评价自己 30 多年写作生涯：“第一个十年我才华横溢，‘贼光闪现’，令周边黯然失色；第二个十年，我终于‘宝光现形’，不再去抢风头，反而与身边的美丽相得益彰；进入第三个十年，繁华落尽见真醇，我进入了‘醇光初现’的阶段，真正体味到了境界之美”。长夜有穷，真水无香。领略过了 K8s“身在江

K8S/Kubernetes

如何基于 K8s 构建下一代 DevOps 平台？

作者 | 孙健波（天元）导读：当前云原生 DevOps 体系现状如何？面临哪些挑战？如何通过 OAM 解决云原生 DevOps 场景下的诸多问题？云原生开发应用模型 OAM(Open Application Model) 社区核心成员孙健波将为大家一一解答，并分享如何基于 OAM 和 Kubernetes 打造无限能力的下一代 DevOps 平台。什么是 DevOps？为什么基于 Kub