无状态与有状态
001 Deployment控制器设计原则：管理的所有Pod一模一样，提供同一个服务，也不考虑在哪台Node运行，可随意 扩容和缩容。这种应用称为“无状态”，例如Web服务
002 在实际的场景中，并不能满足所有应用，尤其是分布式应用，会部署多个实例，这些实例之间往往有依赖关系，例 如主从关系、主备关系，这种应用称为“有状态”，例如MySQL主从、Etcd集群
StatefulSet 控制器概述
StatefulSet控制器用于部署有状态应用，满足一些有状态应用的需求：
001 Pod有序的部署、扩容、删除和停止
002 Pod分配一个稳定的且唯一的网络标识
003 Pod分配一个独享的存储
StatefulSet 控制器：网络标识
稳定的网络标识
001 使用Headless Service（相比普通Service只是将spec.clusterIP定义为None）来维护Pod网络身份，会 为每个Pod分配一个数字编号并且按照编号顺序部署
002 还需要在StatefulSet添加serviceName: “xxx”字段指定StatefulSet控制器要使用这个Headless Service
稳定主要体现在主机名和Pod A记录
001 主机名：<statefulset名称>-<编号>
002 Pod DNS A记录：<statefulset名称-编号>.<service-name> .<namespace>.svc.cluster.local
StatefulSet 控制器：独享存储
001 StatefulSet的存储卷使用VolumeClaimTemplate创建，称为卷申请模板
002 当StatefulSet使用VolumeClaimTemplate创建一个PersistentVolume时，同样也会为每个Pod分配并创 建一个编号的PVC
003 每个PVC绑定对应的PV，从而保证每个Pod都拥有独立的存储
StatefulSet示例
Service
apiVersion: v1
kind: Service
metadata:
name: statefulweb
namespace: default
spec:
clusterIP: None
ports:
- port: 80 # service端口
protocol: TCP # 协议
targetPort: 80 # 容器端口
selector: # 标签选择器
app: nginx # 指定关联Pod的标签
type: ClusterIP # 服务类型
Statefulpod
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: statefulpod
spec:
serviceName: "statefulweb"
replicas: 3
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx
ports:
- containerPort: 80
name: web
volumeMounts:
- name: www
mountPath: /usr/share/nginx/html
volumeClaimTemplates:
- metadata:
name: www
spec:
storageClassName: "managed-nfs-storage"
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 1Gi
-----------------------------------------------------------------------------
先启动StorageClass相关对象，即【基于NFS实现PV动态供给】
[root@k8smaster nfs]# kubectl get pod
NAME READY STATUS RESTARTS AGE
nfs-client-provisioner-5f98b5cdfb-rm7dd 1/1 Running 0 84s
服务启动
#服务按序号依次启动
[root@k8smaster statefulset]# kubectl get pod
NAME READY STATUS RESTARTS AGE
statefulpod-0 1/1 Running 0 85s
statefulpod-1 1/1 Running 0 65s
statefulpod-2 1/1 Running 0 44s
#CLUSTER-IP为None
[root@k8smaster statefulset]# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 24d
statefulweb ClusterIP None <none> 80/TCP 2m32s
验证网络标志
#查看主机名
[root@k8smaster statefulset]# kubectl exec statefulpod-0 -- hostname
statefulpod-0
[root@k8smaster statefulset]# kubectl exec statefulpod-1 -- hostname
statefulpod-1
[root@k8smaster statefulset]# kubectl exec statefulpod-2 -- hostname
statefulpod-2
#测试A记录解析
[root@k8smaster statefulset]# kubectl run -i -t dns-test --rm --image busybox:1.28.4 -- sh
If you don't see a command prompt, try pressing enter.
/ # nslookup statefulweb
Server: 10.96.0.10
Address 1: 10.96.0.10 kube-dns.kube-system.svc.cluster.local
Name: statefulweb
Address 1: 10.244.249.41 statefulpod-0.statefulweb.default.svc.cluster.local
Address 2: 10.244.249.60 statefulpod-2.statefulweb.default.svc.cluster.local
Address 3: 10.244.249.43 statefulpod-1.statefulweb.default.svc.cluster.local
/ # ping statefulpod-0.statefulweb.default.svc.cluster.local
PING statefulpod-0.statefulweb.default.svc.cluster.local (10.244.249.41): 56 data bytes
64 bytes from 10.244.249.41: seq=0 ttl=63 time=0.105 ms
64 bytes from 10.244.249.41: seq=1 ttl=63 time=0.162 ms
/ # ping statefulpod-1.statefulweb.default.svc.cluster.local
PING statefulpod-1.statefulweb.default.svc.cluster.local (10.244.249.43): 56 data bytes
64 bytes from 10.244.249.43: seq=0 ttl=63 time=0.087 ms
64 bytes from 10.244.249.43: seq=1 ttl=63 time=0.090 ms
/ # ping statefulpod-2.statefulweb.default.svc.cluster.local
PING statefulpod-2.statefulweb.default.svc.cluster.local (10.244.249.60): 56 data bytes
64 bytes from 10.244.249.60: seq=0 ttl=63 time=0.081 ms
64 bytes from 10.244.249.60: seq=1 ttl=63 time=0.115 ms
验证独立存储
[root@k8smaster statefulset]# kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS
www-statefulpod-0 Bound pvc-1d88760c-8207-4e6e-9f81-520cce33051f 1Gi RWO managed-nfs-storage
www-statefulpod-1 Bound pvc-bd7d64f7-31af-4484-a5e4-d7d3bc8ecf2f 1Gi RWO managed-nfs-storage
www-statefulpod-2 Bound pvc-6256144d-7dac-4162-8d85-fda6e128b899 1Gi RWO managed-nfs-storage
[root@k8smaster statefulset]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON
pvc-1d88760c-8207-4e6e-9f81-520cce33051f 1Gi RWO Delete Bound default/www-statefulpod-0 managed-nfs-storage
pvc-6256144d-7dac-4162-8d85-fda6e128b899 1Gi RWO Delete Bound default/www-statefulpod-2 managed-nfs-storage
pvc-bd7d64f7-31af-4484-a5e4-d7d3bc8ecf2f 1Gi RWO Delete Bound default/www-statefulpod-1 managed-nfs-storage
[root@k8smaster statefulset]# kubectl get pod -o wide
statefulpod-0 1/1 Running 10.244.249.41 k8snode1
statefulpod-1 1/1 Running 10.244.249.43 k8snode1
statefulpod-2 1/1 Running 10.244.249.60 k8snode1
[root@k8smaster statefulset]# cd /ifs/kubernetes/
#在对应的3个目录下分别编辑index.html
[root@k8smaster statefulset]# curl 10.244.249.41
hello statefulpod-0
[root@k8smaster statefulset]# curl 10.244.249.43
hello statefulpod-1
[root@k8smaster statefulset]# curl 10.244.249.60
hello statefulpod-2
StatefulSet 控制器：小结
StatefulSet与Deployment区别：有身份的！
身份三要素：
001 域名
002 主机名
003 存储（PVC）
网友评论