k8s资源清单描述

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
ApiVersion:
	资源的版本，官方对每个资源规定了不同的版本，我们需要根据资源的版本进行书写。
	一般情况下，资源的版本很少变动，除非在升级过程中，随着功能的增强，版本可能也会微调。

kind:
	资源的类型，K8S集群中一切皆资源。
	
metadata:
	资源的元数据信息，包括但不限于： 
		- 资源的名称;(常用)
		- 资源所属的名称空间;(常用)
		- 资源的标签;(常用)
		- 资源的注解;(早期版本常用)
		- 资源的创建时间;
		- ...

spec:
	定义了用户对Pod的期望状态。用户可以做如下操作:
		- 指定节点调度;
		- 指定调度的亲和性或反亲和性;
		- 资源的数据是否进行持久化;
		- 资源的配置文件是否需要修改;
		- 资源的容器使用具体的镜像;
		- 资源的限制
		- 资源的认证信息
		- 资源的安全信息
		- ...

status:
	资源实际的状态，比如调度到哪个worker节点，实际分配的IP地址，运行周期，重启次数等。
	该资源一般不需要人为定义，而是有K8S集群自身维护。
	
[root@master231 ~]# kubectl api-resources
查看所有资源

pod资源及资源清单

1.POD资源

kubernetes资源对象操作

资源对象 增 删 改 查
node kubectl delete node kubectl get nodes
namespace kubectl create ns [资源名] kubectl delete ns [资源名] kubectl edit ns [资源名] kubectl get ns
pod kubectl delete pod [资源名] kubectl edit pod[资源名] kubectl get pod

查看集群节点状态

1
2
3
4
5
6
[root@master ~]# kubectl get nodes
NAME     STATUS   ROLES    AGE     VERSION
master   Ready    master   4d13h   v1.19.3
node01   Ready    <none>   4d13h   v1.19.3
node02   Ready    <none>   4d13h   v1.19.3
node03   Ready    <none>   4d13h   v1.19.3

查看名称空间

1
2
3
4
5
6
7
8
[root@master ~]# kubectl get ns
NAME              STATUS   AGE
default           Active   4d13h
kube-flannel      Active   4d12h
kube-node-lease   Active   4d13h
kube-public       Active   4d13h
kube-system       Active   4d13h
luay              Active   4h17m

查看pod (默认default名称空间)

1
2
3
4
[root@master ~]# kubectl get pod
NAME                     READY   STATUS    RESTARTS   AGE
nginx-565785f75c-tpjmj   1/1     Running   0          6h8m
nginx-pod                1/1     Running   0          5h43m

查看pod (指定名称空间)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
[root@master ~]# kubectl get pods -n luay
NAME          READY   STATUS    RESTARTS   AGE
busybox-pod   2/2     Running   0          3m30s


[root@master ~]# kubectl get pod -n luay -o wide
NAME          READY   STATUS    RESTARTS   AGE     IP         NODE     NOMINATED NODE   READINESS GATES
busybox-pod   2/2     Running   0          4m14s   10.2.2.6   node03   <none>           <none>

-n 指定命名空间
-o 指定输出方式
json：输出json格式
yaml：输出yaml格式
wide：输出详细信息

2.pod的资源清单

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
[root@master01 ~]# vim nginx.yaml
apiVersion: v1
kind: Pod
metadata:
  name: nginx-pod
  namespace: default
spec:
  containers:
  - image: nginx:alpine
    imagePullPolicy: IfNotPresent
    name: nginx-containers

`注释`
apiVersion: v1       # K8S资源接口
kind: Pod            # 资源类型
metadata:            #资源的元数据
  name: nginx-pod    # POD名字
  namespace: default # 指定资源启动在哪个名称空间，默认default
spec:                #容器相关信息
  containers:        #容器信息
  - image: nginx:alpine           #指定容器镜像
    imagePullPolicy: IfNotPresent #指定镜像拉取规则
    name: nginx-containers        #容器名字


Always：总是拉取镜像，不管当前机器上是否有该镜像都拉取 nginx:latest
Never：从不拉取镜像，需要提前docker pull
IfNotPresent：如果镜像不存在，则拉取

资源清单的增，删，查

1.编写资源清单

1
2
3
4
5
6
7
8
9
[root@master231:1 pods]# cat single-pods.yaml 
apiVersion: v1  # 指定资源的api版本信息
kind: Pod       # 指定资源的类型
metadata:       # 指定Pod的元数据信息
  name: xiuxian # 指的Pod的名称
spec:           # 定义期望资源
  containers:   # 定义容器资源
  - name: c1    # 容器的名称
    image: registry.cn-hangzhou.aliyuncs.com/yinzhengjie-k8s/apps:v1  # 容器的镜像

2.创建资源

1
2
[root@master231:1 pods]# kubectl create -f single-pods.yaml 
pod/xiuxian created

3.查看资源

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
[root@master231:1 pods]# kubectl get pods
NAME      READY   STATUS    RESTARTS   AGE
xiuxian   1/1     Running   0          54s

[root@master231:1 pods]# kubectl get pods -o wide
NAME      READY   STATUS    RESTARTS   AGE   IP           NODE        NOMINATED NODE   READINESS GATES
xiuxian   1/1     Running   0          86s   10.100.2.4   worker233   <none>           <none>

[root@master231:1 pods]# curl 10.0.100.2.4
curl: (6) Could not resolve host: 10.0.100.2.4
[root@master231:1 pods]# curl 10.100.2.4
<!DOCTYPE html>
<html>
  <head>
    <meta charset="utf-8"/>
    <title>yinzhengjie apps v1</title>
    <style>
       div img {
          width: 900px;
          height: 600px;
          margin: 0;
       }
    </style>
  </head>

  <body>
    <h1 style="color: green">凡人修仙传 v1 </h1>
    <div>
      <img src="1.jpg">
    <div>
  </body>

</html>

4.删除资源

1
2
3
4
5
6
[root@master231:1 pods]# kubectl delete -f single-pods.yaml 
pod "xiuxian" deleted

[root@master231:1 pods]# kubectl get pods -o wide
No resources found in default namespace.
[root@master231:1 pods]#

5.使用apply命令创建或者修改资源

1
2
3
4
5
6
7
8
9
10
#第一次执行则是创建
[root@master231:1 pods]# kubectl apply -f single-pods.yaml 
pod/xiuxian created

[root@master231:1 pods]# kubectl get pods -o wide
NAME      READY   STATUS    RESTARTS   AGE   IP           NODE        NOMINATED NODE   READINESS GATES
xiuxian   1/1     Running   0          23s   10.100.2.5   worker233   <none>           <none>
#再次执行则是修改资源，若没有修改资源清单则提示unchanged
[root@master231:1 pods]# kubectl apply -f single-pods.yaml 
pod/xiuxian unchanged

响应式·基于标签管理POD资源

1.查看资源的标签

1
2
3
[root@master231:1 pods]# kubectl get pods --show-labels
NAME      READY   STATUS    RESTARTS   AGE     LABELS
xiuxian   1/1     Running   0          5m54s   <none>

2.给资源打标签

1
2
3
4
5
[root@master231:1 pods]# kubectl label pod xiuxian name=wanglin level=huashen
pod/xiuxian labeled
[root@master231:1 pods]# kubectl get pods --show-labels
NAME      READY   STATUS    RESTARTS   AGE     LABELS
xiuxian   1/1     Running   0          9m17s   level=huashen,name=wanglin

3.将资源移除标签

1
2
3
4
5
6
[root@master231:1 pods]# kubectl label pod xiuxian name-
pod/xiuxian unlabeled

[root@master231:1 pods]# kubectl get pods --show-labels
NAME      READY   STATUS    RESTARTS   AGE   LABELS
xiuxian   1/1     Running   0          10m   level=huashen

4.修改标签 

1
2
3
4
5
6
7
8
9
10
11
[root@master231:1 pods]# kubectl get pods --show-labels
NAME      READY   STATUS    RESTARTS   AGE   LABELS
xiuxian   1/1     Running   0          10m   level=huashen

[root@master231:1 pods]# kubectl label pod xiuxian level=jindan --overwrite
pod/xiuxian labeled
[root@master231:1 pods]# kubectl get pods --show-labels
NAME      READY   STATUS    RESTARTS   AGE   LABELS
xiuxian   1/1     Running   0          12m   level=jindan

`--overwrite`不加上则报错level的值已存在

5.为所有Pod资源打标签

1
2
3
4
5
6
7
8
[root@master231:1 pods]# kubectl label pods --all app=xiuxian
pod/xiuxian labeled
pod/xiuxian-v1 labeled

[root@master231:1 pods]# kubectl get pods --show-labels
NAME         READY   STATUS    RESTARTS   AGE    LABELS
xiuxian      1/1     Running   0          20m    app=xiuxian,level=jindan
xiuxian-v1   1/1     Running   0          101s   app=xiuxian

基于标签过滤资源

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
[root@master231:1 pods]# kubectl get pods --show-labels
NAME         READY   STATUS    RESTARTS   AGE     LABELS
xiuxian      1/1     Running   0          25m     app=xiuxian,level=jindan
xiuxian-v1   1/1     Running   0          6m41s   app=xiuxian,name=wanglin

#过滤key里含有name的标签
[root@master231:1 pods]# kubectl get pods --show-labels -l name
NAME         READY   STATUS    RESTARTS   AGE     LABELS
xiuxian-v1   1/1     Running   0          6m21s   app=xiuxian,name=wanglin

#过滤level=jindan的标签
[root@master231:1 pods]# kubectl get pods --show-labels -l level=jindan
NAME      READY   STATUS    RESTARTS   AGE   LABELS
xiuxian   1/1     Running   0          26m   app=xiuxian,level=jindan

#过滤level不等于jindan的标签
[root@master231:1 pods]# kubectl get pods --show-labels -l  level!=jindan
NAME         READY   STATUS    RESTARTS   AGE     LABELS
xiuxian-v1   1/1     Running   0          8m54s   app=xiuxian,name=wanglin

7.基于标签删除资源

1
2
3
4
5
6
7
8
9
10
11
[root@master231:1 pods]# kubectl get pods --show-labels
NAME         READY   STATUS    RESTARTS   AGE   LABELS
xiuxian      1/1     Running   0          30m   app=xiuxian,level=jindan
xiuxian-v1   1/1     Running   0          11m   app=xiuxian,name=wanglin

[root@master231:1 pods]# kubectl delete pods -l name
pod "xiuxian-v1" deleted

[root@master231:1 pods]# kubectl get pods --show-labels
NAME      READY   STATUS    RESTARTS   AGE   LABELS
xiuxian   1/1     Running   0          31m   app=xiuxian,level=jindan

声明式·修改资源标签

1.修改资源清单 记得修改后要应用

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
[root@master231:1 pods]# cat single-pods.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: xiuxian
  # 为资源打标签
  labels:
    name: wanglin
    level: jindan
spec:
  containers:
  - name: c1
    image: registry.cn-hangzhou.aliyuncs.com/yinzhengjie-k8s/apps:v1

#资源清单修改后要应用一下
[root@master231:1 pods]# kubectl apply -f single-pods.yaml 
pod/xiuxian created

2.查看资源清单

1
2
3
[root@master231:1 pods]# kubectl get pods --show-labels
NAME      READY   STATUS    RESTARTS   AGE   LABELS
xiuxian   1/1     Running   0          76s   level=jindan,name=wanglin

修改标签和删除标签直接在资源清单里修改即可，修改后应用一下

响应式和声明式的区别

1.相同点

​ 都可以管理资源

2.不同点

  • 响应式修改后立即生效，但是资源被重新创建后又得重新执行相关命令;
  • 声明式修改需要先修改资源清单，在使用apply命令更新应用配置;

连接容器

1.连接指定pod执行命令

1
2
3
4
5
kubectl exec -it 资源名 command
[root@master231:0 kubernetes]# kubectl exec -it mysql80 -- mysql
···
Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.
mysql>

2.连接同一个Pod的不同容器执行命令

1
2
3
4
5
6
[root@master231 pods]# kubectl exec -it multiple-wordpress -c db -- ls / -l

温馨提示:
	1.kubectl exec和docker exec都是在容器中执行命令;
	2.如果一个Pod有多个容器，则可以使用-c连接到指定的容器执行相应的命令;
	3.如果一个Pod有多个容器，若不使用-c连接到指定容器执行命令，则默认连接到第一个容器;

案例

基于pod部署mysql数据库

1.编写资源清单

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
[root@master231:1 pods]# cat single-pods-mysql.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: mysql80
  labels:
    app: mysql
spec:
   # 指定调度到哪个节点，该节点必须在etcd数据库中有记录
  nodeName: worker233
  containers:
  - name: mysqld
    image: mysql:8.0
    env:
      # 指定变量的名称,允许密码为空
    - name: MYSQL_ALLOW_EMPTY_PASSWORD
      # 指定变量的值
      value: "yes"
    - name: MYSQL_DATABASE
      value: wordpress
    - name: MYSQL_USER
      value: wp_usre
    - name: MYSQL_PASSWORD
      value: "123"

2.运行msyql实例

1
2
[root@master231:0 kubernetes]# kubectl apply -f single-pods-mysql.yaml 
pod/mysql80 created

3.查看pod信息

1
2
3
4
5
[root@master231:0 kubernetes]# kubectl get pod -o wide
NAME      READY   STATUS             RESTARTS   AGE   IP            NODE     
mysql80   0/1     ImagePullBackOff   0          83s   10.100.2.12   worker233

#创建节点worker233   ip 10.100.2.12

基于一个pod部署wordpress

1.编写资源清单

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
[root@master kubernetes]# cat wordpress.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: wordpress-pod
spec:
 # 使用宿主机网络名称空间
  hostNetwork: true
  containers:
  - name: db
    image: mysql:8.0
    imagePullPolicy: IfNotPresent
    env:
    - name: MYSQL_ALLOW_EMPTY_PASSWORD
      value: "yes"
    - name: MYSQL_DATABASE
      value: wp
    - name: MYSQL_USER
      value: wp_usre
    - name: MYSQL_PASSWORD
      value: '123'
  - name: wp
    image: wordpress:latest
    imagePullPolicy: IfNotPresent
    env:
    - name: WORDPRESS_DB_HOST
      value: "127.0.0.1"
    - name: WORDPRESS_DB_NAME
      value: wp
    - name: WORDPRESS_DB_USER
      value: wp_usre
    - name: WORDPRESS_DB_PASSWORD
      value: '123'

`保证宿主机80端口不被占用`

2.根据资源清单创建pod

1
[root@master kubernetes]# kubectl apply -f wordpress.yaml

3.查看pod信息

1
2
3
4
5
6
7
8
[root@master kubernetes]# kubectl get pod  -o wide
NAME                     READY   STATUS    RESTARTS   AGE   IP           NODE     NOMINATED NODE   READINESS GATES
nginx-565785f75c-tpjmj   1/1     Running   0          8h    10.2.2.2     node03   <none>           <none>
nginx-pod                1/1     Running   0          8h    10.2.2.3     node03   <none>           <none>
wordpress-pod            2/2     Running   0          26m   10.0.0.203   node03   <none>           <none>

#pod创建节点ip为10.0.0.203
浏览器访问：100.0.203