基于StatefulSet有状态部署与DaemonSet部署prometheus node-exporter
statefulset有状态应用部署
考虑无状态的pod因素，让每个pod独立，保持pod启动顺序和唯一性。拥有唯一的网络标识符，持久存储。启动顺序有序，例如mysql主从节点
应用场景
为了解决有服务状态的集群部署、集群之间的数据同步问题（Mysql主从等）
稳定的持久化存储：Pod重新调度后访问相同的持久化数据，基于PVC来实现。
稳定的网络标志：Pod重新调度后PodName和HostName不变，基于Headless Service来实现。
有序扩展：在扩展的时候要依据定义的顺序依次依次进行，基于init containers 实现。在下一个Pod运行之前所有Pod必须是Running和Ready状态）。
有序删除，从后到前逐个删除pod。
组成部分
从上面的应用场景可以发现，StatefulSet由以下几个部分组成：
Headless Service，用于定义网络标志（DNS domain）的。
volumeClaimTemplates，用于创建PVC，指定pvc名称大小，pvc必须由存储类提供存储。为每个Pod生成不同的pvc，并绑定pv，从而实现各pod有专用存储。
StatefulSet，定义具体应用，与deployments类似。
deployment和statefulset区别：有身份（具有唯一标识符）
（1）根据主机名+按照一定规则生成域名
（2）每个pod拥有唯一的主机名
（3）唯一域名：格式：
（4）Pod主机名称.service名称.名称空间.svc.cluster.local
部署有状态服务：
声明svc取消ip地址的暴露，将clusterip置位None
root@deploy:/statefulset# kubectl create ns test
namespace/test created
root@deploy:/statefulset# vim statefulset.yaml
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: web-statefulset
namespace: test
spec:
serviceName: test-statefulset
replicas: 2
selector:
matchLabels:
app: web
template:
metadata:
labels:
app: web
spec:
containers:
- image: nginx
name: nginx
ports:
- containerPort: 80
- name: tomcat
image: tomcat
ports:
- containerPort: 8080
---
apiVersion: v1
kind: Service
metadata:
labels:
app: web
name: test-statefulset
namespace: test
spec:
clusterIP: None
ports:
- port: 80
name: nginx
protocol: TCP
targetPort: 80
- port: 8080
protocol: TCP
targetPort: 8080
name: tomcat
selector:
app: web
root@deploy:/statefulset# kubectl apply -f statefulset.yaml
查看statefulset pod
root@deploy:/statefulset# kubectl get pods
NAME READY STATUS RESTARTS AGE
web-statefulset-0 2/2 Running 0 70s
web-statefulset-1 2/2 Running 0 38s
进入到pod测试网络和主机名称
#查看svc
root@deploy:/statefulset# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
test-statefulset ClusterIP None <none> 80/TCP,8080/TCP 8m33s
root@deploy:/statefulset# kubectl exec -it web-statefulset-0 -- bash
Defaulted container "nginx" out of: nginx, tomcat
root@web-statefulset-0:/# cat /etc/hosts
# Kubernetes-managed hosts file.
127.0.0.1 localhost
::1 localhost ip6-localhost ip6-loopback
fe00::0 ip6-localnet
fe00::0 ip6-mcastprefix
fe00::1 ip6-allnodes
fe00::2 ip6-allrouters
10.200.166.175 web-statefulset-0.test-statefulset.test.svc.cluster.local web-statefulset-0
使用其他namespace下的pod访问test 名称空间下的statefulset应用
root@master1:~# kubectl get pods -n default
NAME READY STATUS RESTARTS AGE
aliyun-web-deployment-668cb5f7b5-5rnqs 1/1 Running 1 (127m ago) 3d21h
aliyun-web-deployment-668cb5f7b5-sgs2g 1/1 Running 1 (127m ago) 3d21h
web-deployment-647db54fb5-fkggx 1/1 Running 1 (127m ago) 4d1h
web-deployment-647db54fb5-lz64s 1/1 Running 1 (127m ago) 4d1h
root@master1:~# kubectl exec -it -n default web-deployment-647db54fb5-lz64s -- curl test-statefulset.test
<!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>
<body>
<h1>Welcome to nginx!</h1>
<p>If you see this page, the nginx web server is successfully installed and
working. Further configuration is required.</p>
<p>For online documentation and support please refer to
<a href="http://nginx.org/">nginx.org</a>.<br/>
Commercial support is available at
<a href="http://nginx.com/">nginx.com</a>.</p>
<p><em>Thank you for using nginx.</em></p>
</body>
</html>
root@master1:~#
pod访问测试
root@web-statefulset-0:/usr/local/tomcat# curl web-statefulset-0.test-statefulset.test.svc.cluster.local
<!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>
<body>
<h1>Welcome to nginx!</h1>
<p>If you see this page, the nginx web server is successfully installed and
working. Further configuration is required.</p>
<p>For online documentation and support please refer to
<a href="http://nginx.org/">nginx.org</a>.<br/>
Commercial support is available at
<a href="http://nginx.com/">nginx.com</a>.</p>
<p><em>Thank you for using nginx.</em></p>
</body>
</html>
DaemonSet守护进程
DaemonSet确保集群中每个（部分）node运行一份pod副本，当node加入集群时创建pod，当node离开集群时回收pod。如果删除DaemonSet，其创建的所有pod也被删除，DaemonSet中的pod覆盖整个集群。
DaemonSet配置文件的语法和结构与Deployment几乎完全一样，区别是：
1.kind设为DaemonSet
2.不需replicas
DaemonSet 的一些典型使用场景：
在每个节点上运行集群守护进程
在每个节点上运行日志收集守护进程
在每个节点上运行监控守护进程
设置maser节点污点容忍，在master节点部署pod
root@deploy:/daemonset# kubectl get nodes
NAME STATUS ROLES AGE VERSION
192.168.100.2 Ready,SchedulingDisabled master 11d v1.24.3
192.168.100.5 Ready node 11d v1.24.3
192.168.100.6 Ready node 11d v1.24.3
查看master污点
root@deploy:/daemonset# kubectl get nodes
NAME STATUS ROLES AGE VERSION
192.168.100.2 Ready,SchedulingDisabled master 11d v1.24.3
192.168.100.5 Ready node 11d v1.24.3
192.168.100.6 Ready node 11d v1.24.3
root@deploy:/daemonset# kubectl describe node 192.168.100.2|grep -i taint
Taints: node.kubernetes.io/unschedulable:NoSchedule
创建daemonset，污点容忍key填写master设置的污点key
设置 hostNetwork: true，将pod端口映射在k8s节点
root@deploy:/daemonset# vim daemonset.yaml
apiVersion: apps/v1
kind: DaemonSet
metadata:
labels:
app: web
name: web
namespace: test
spec:
selector:
matchLabels:
app: web
template:
metadata:
labels:
app: web
spec:
hostNetwork: true
hostPID: true
containers:
- image: nginx
name: nginx
imagePullPolicy: IfNotPresent
ports:
- containerPort: 80
tolerations:
- key: "node.kubernetes.io/unschedulable"
operator: "Exists"
effect: "NoSchedule"
查看k8s节点端口80监听
root@master1:~# lsof -i :80
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
nginx 114536 root 7u IPv4 569545 0t0 TCP *:http (LISTEN)
nginx 114536 root 8u IPv6 569546 0t0 TCP *:http (LISTEN)
nginx 114583 systemd-resolve 7u IPv4 569545 0t0 TCP *:http (LISTEN)
nginx 114583 systemd-resolve 8u IPv6 569546 0t0 TCP *:http (LISTEN)
nginx 114583 systemd-resolve 13u IPv4 589033 0t0 TCP master1:http->192.168.100.1:60148 (ESTABLISHED)
nginx 114584 systemd-resolve 7u IPv4 569545 0t0 TCP *:http (LISTEN)
nginx 114584 systemd-resolve 8u IPv6 569546 0t0 TCP *:http (LISTEN)
客户端访问：
部署普罗米修斯node客户端
设置hostNetwork: true将pod端口在k8s节点监听；
设置hostPort: 9100，将pod端口监听在宿主机k8s节点的9100端口
root@deploy:/daemonset# vim prometheus_node.yaml
apiVersion: apps/v1
kind: DaemonSet
metadata:
labels:
app: prometheus-node
name: prometheus-node-daemonset
namespace: test
spec:
selector:
matchLabels:
app: prometheus-node
template:
metadata:
labels:
app: prometheus-node
spec:
hostNetwork: true
hostPID: true
tolerations:
- key: "node.kubernetes.io/unschedulable"
operator: "Exists"
effect: "NoSchedule"
containers:
- image: prom/node-exporter:v1.3.1
name: prometheus-node
imagePullPolicy: IfNotPresent
ports:
- containerPort: 9100
hostPort: 9100
protocol: TCP
name: metrics
volumeMounts:
- mountPath: /host/proc
name: proc
- mountPath: /host/sys
name: sys
- mountPath: /host
name: rootfs
args:
- --path.procfs=/host/proc
- --path.sysfs=/host/sys
- --path.rootfs=/host
volumes:
- name: proc
hostPath:
path: /proc
- name: sys
hostPath:
path: /sys
- name: rootfs
hostPath:
path: /
查看pod在集群中运行
客户端访问：节点端口
master节点
node1节点
node2节点
本文来自博客园，作者：PunchLinux，转载请注明原文链接：https://www.cnblogs.com/punchlinux/p/16561471.html
