k8s——statefulset

statefulset基础模版

[root@master statefulset]# cat web.yaml
---
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     # statefulset 类型的资源
metadata:
  name: web   # statefulset 对象的名字
spec:
  serviceName: "nginx"   # 使用哪个service来管理dns
  replicas: 2
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.7.9
        ports:   # 容器内部要暴露的端口
        - containerPort: 80  # 具体暴露的端口号
          name: web  #该端口配置的名字

// 先不写下面的，因为涉及到别的知识点

        volumeMounts:   # 加载数据卷
        - name: www    #指定加载哪个数据卷
          mountPath: /usr/share/nginx/html  # 加载到容器中的哪个目录
  volumeClaimTemplates:    # 数据卷模版
  - metadata:    # 数据卷描述
      name: www   # 数据卷的名称
      annotations:   # 数据卷的注解
        volume.alpha.kubernetes,io.storage-class: anything
    spec:   # 数据卷的规约
      accessModes: [ "ReadWriteOnce" ]   # 访问模式
      resources:
        requests:
          storage: 1Gi   # 需要1G 的存储资源
[root@master statefulset]# 

[root@master statefulset]# kubectl create -f web.yaml
service/nginx created
statefulset.apps/web created
[root@master statefulset]# kubectl get sts
NAME   READY   AGE
web    0/2     5s
[root@master statefulset]# kubectl get svc
NAME         TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   5d18h
nginx        ClusterIP   None         <none>        80/TCP    8s
[root@master statefulset]# kubectl get pvc  //注释掉之后这个是没有的
NAME        STATUS    VOLUME   CAPACITY   ACCESS MODES   STORAGECLASS   AGE
www-web-0   Pending                                                     11s

//修改后
[root@master statefulset]# kubectl create -f web.yaml
service/nginx created
statefulset.apps/web created
[root@master statefulset]# kubectl get  sts
NAME   READY   AGE
web    2/2     9s
[root@master statefulset]# kubectl get  svc
NAME         TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   5d19h
nginx        ClusterIP   None         <none>        80/TCP    15s
[root@master statefulset]#
#  有一个疑问，k8s集群内部`kubectl get po`并没有这么多pod
[root@master ~]# kubectl get  po
NAME         READY   STATUS    RESTARTS       AGE
my-pod       1/1     Running   3 (4m8s ago)   5d18h
nginx-demo   1/1     Running   4 (4m8s ago)   5d23h
nginx-po     1/1     Running   4 (4m8s ago)   5d22h
web-0        1/1     Running   2 (4m8s ago)   20h
web-1        1/1     Running   2 (4m8s ago)   20h
[root@master ~]# 

# python 中列出pod ，列出的是所有的pod 等价于kubectl get -A po
[root@master ~]# kubectl get -A po
NAMESPACE         NAME                               READY   STATUS    RESTARTS        AGE
default           my-pod                             1/1     Running   3 (3m12s ago)   5d18h
default           nginx-demo                         1/1     Running   4 (3m12s ago)   5d23h
default           nginx-po                           1/1     Running   4 (3m12s ago)   5d22h
default           web-0                              1/1     Running   2 (3m12s ago)   20h
default           web-1                              1/1     Running   2 (3m12s ago)   20h
kube-flannel      kube-flannel-ds-bkvjm              1/1     Running   4 (3m12s ago)   5d23h
kube-flannel      kube-flannel-ds-hg6bh              1/1     Running   5 (3m56s ago)   6d14h
kube-system       coredns-66f779496c-c5rmz           1/1     Running   4 (3m56s ago)   6d15h
kube-system       coredns-66f779496c-jswxz           1/1     Running   4 (3m56s ago)   6d15h
kube-system       etcd-master                        1/1     Running   4 (3m56s ago)   6d15h
kube-system       kube-apiserver-master              1/1     Running   4 (3m56s ago)   6d15h
kube-system       kube-controller-manager-master     1/1     Running   5 (3m56s ago)   6d15h
kube-system       kube-proxy-jr5m5                   1/1     Running   4 (3m56s ago)   6d15h
kube-system       kube-proxy-tcpzh                   1/1     Running   4 (3m12s ago)   5d23h
kube-system       kube-scheduler-master              1/1     Running   5 (3m57s ago)   6d15h
tigera-operator   tigera-operator-55585899bf-s28bt   1/1     Running   8 (3m19s ago)   6d14h

拉取指定镜像测试

[root@master statefulset]# kubectl run -it --image busybox:1.28.4 dns-test  /bin/sh
If you don't see a command prompt, try pressing enter.
/ # ping web-0.nginx
PING web-0.nginx (10.244.1.45): 56 data bytes
64 bytes from 10.244.1.45: seq=0 ttl=64 time=0.048 ms
64 bytes from 10.244.1.45: seq=1 ttl=64 time=0.054 ms
^C
--- web-0.nginx ping statistics ---
2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 0.048/0.051/0.054 ms
/ # nslookup web-0.nginx    # sts的名字-第几个.service的名字 是statefulset的访问格式
Server:    10.96.0.10
Address 1: 10.96.0.10 kube-dns.kube-system.svc.cluster.local

Name:      web-0.nginx
Address 1: 10.244.1.45 web-0.nginx.default.svc.cluster.local
/ #

扩容和缩容

# 扩容
[root@master statefulset]# kubectl scale sts web --replicas=6
statefulset.apps/web scaled

# 缩容
[root@master statefulset]# kubectl scale sts web --replicas=3
statefulset.apps/web scaled
[root@master statefulset]# kubectl get sts
NAME   READY   AGE
web    3/3     20h
[root@master statefulset]#

灰度更新

[root@master statefulset]# kubectl scale sts web --replicas=6
statefulset.apps/web scaled
[root@master statefulset]# kubectl get sts
NAME   READY   AGE
web    6/6     20h

[root@master statefulset]# kubectl  edit sts web
statefulset.apps/web edited
# 进去之后修改
---
  updateStrategy:
    rollingUpdate:
      partition: 3   # 把0修改为3
    # 修改partition的值可以模拟灰度更新，就是说更新的时候先更新编号大于3的pod
---

# 查看
[root@master statefulset]# kubectl describe po web-1|grep nginx:
  nginx:
    Image:          nginx:1.7.9
  Normal   Pulled                  20h                kubelet            Container image "nginx:1.7.9" already present on machine
  Normal   Pulled                  12h                kubelet            Container image "nginx:1.7.9" already present on machine
  Normal   Pulled                  48m                kubelet            Container image "nginx:1.7.9" already present on machine
[root@master statefulset]# kubectl describe po web-5|grep nginx:
  nginx:
    Image:          nginx:1.9.1
  Normal  Pulled     95s   kubelet            Container image "nginx:1.9.1" already present on machine
[root@master statefulset]# 

# 确认新版本没有问题之后，在修改partition的值为0，来更新所有
[root@master statefulset]# kubectl  edit sts web
statefulset.apps/web edited
[root@master statefulset]# kubectl describe po web-5|grep nginx:
  nginx:
    Image:          nginx:1.9.1
  Normal  Pulled     6m13s  kubelet            Container image "nginx:1.9.1" already present on machine
[root@master statefulset]# kubectl describe po web-0|grep nginx:
  nginx:
    Image:          nginx:1.9.1
  Normal  Pulled     4s    kubelet            Container image "nginx:1.9.1" already present on machine
[root@master statefulset]# 

OnDelete策略更新

[root@master statefulset]# kubectl edit sts web
statefulset.apps/web edited
---
  updateStrategy:
    type: OnDelete   #修改更新策略
---

[root@master statefulset]# kubectl edit  sts web
statefulset.apps/web edited
---
    spec:
      containers:
      - image: nginx:1.7.9  #修改镜像
        imagePullPolicy: IfNotPresent
        name: nginx
        ports:
---
[root@master statefulset]# kubectl describe po web-0 |grep nginx:
  nginx:
    Image:          nginx:1.9.1
  Normal  Pulled     13m   kubelet            Container image "nginx:1.9.1" already present on machine
[root@master statefulset]# kubectl describe po web-1 |grep nginx:
  nginx:
    Image:          nginx:1.9.1
  Normal  Pulled     14m   kubelet            Container image "nginx:1.9.1" already present on machine
[root@master statefulset]# kubectl delete pod web-1
pod "web-1" deleted  # 看似删除实则更新
[root@master statefulset]#
[root@master statefulset]# kubectl describe po web-1 |grep nginx:
  nginx:
    Image:          nginx:1.7.9
  Normal  Pulled     3s    kubelet            Container image "nginx:1.7.9" already present on machine
[root@master statefulset]# 

删除sts

rs是deployment的东西，所以sts没有rs

删除sts和headless service的时候分为两种，级联删除和非级联删除

级联删除

# 删除sts时候，会同时删除pods
[root@master statefulset]# kubectl get sts
NAME   READY   AGE
web    3/3     23h
[root@master statefulset]# kubectl delete sts web
statefulset.apps "web" deleted
[root@master statefulset]# kubectl get sts
No resources found in default namespace.
[root@master statefulset]# kubectl get po
NAME       READY   STATUS    RESTARTS       AGE
dns-test   1/1     Running   1 (160m ago)   162m
[root@master statefulset]# 

# 删除sts的时候，直接删除pod

非级联删除

[root@master statefulset]# kubectl get sts
NAME   READY   AGE
web    2/2     25s
[root@master statefulset]# kubectl get po
NAME       READY   STATUS    RESTARTS       AGE
dns-test   1/1     Running   1 (166m ago)   168m
web-0      1/1     Running   0              27s
web-1      1/1     Running   0              26s
[root@master statefulset]# kubectl delete  sts web --cascade=false
warning: --cascade=false is deprecated (boolean value) and can be replaced with --cascade=orphan.
statefulset.apps "web" deleted
[root@master statefulset]# kubectl get po
NAME       READY   STATUS    RESTARTS       AGE
dns-test   1/1     Running   1 (166m ago)   168m
web-0      1/1     Running   0              74s
web-1      1/1     Running   0              73s
[root@master statefulset]# kubectl get sts
No resources found in default namespace.
[root@master statefulset]#
# 非级联删除没有连带删除pod