容器基础(十): 使用kubernetes部署应用

概述

使用之前的脚本(env/server.py 得到 env/server:v0.1 镜像, env/worker.py 得到 env/worker:v0.1)得到的镜像，在部署好kubernetes的机器上进行测试。kubernetes-v1.12.2当前只部署在master上，且master的Taint已经被删除！基本的做法是：

. 配置一个server-rc.yml, 用于创建server端，全局只配置一个replicas;
. 配置一个server-svc.yml, 与server-rc.yml进行关联, 为server端分配VIP;
. 配置一个worker-rc.yml, 用于创建客户端, 可配置多个replicas, 用于和server进行通信;

YAML脚本

 ➜  k8s  cat server-rc.yml
 apiVersion: v1
 kind: ReplicationController
 metadata:
   name: server
 spec:
   replicas: 1
   selector:
     app: server
   template:
     metadata:
       labels:
         app: server
     spec:
       containers:
         - name: server
           image: env/server:v0.1
           ports:
           - containerPort: 3000
           env:
           - name: APP_PORT
             value: "3000"
 ➜  k8s  

 ➜  k8s  cat server-svc.yml
 apiVersion: v1
 kind: Service
 metadata:
   name: server
 spec:
   ports:
     - port: 3000
   selector:
     app: server
 ➜  k8s  

 ➜  k8s  cat worker-rc.yml
 apiVersion: v1
 kind: ReplicationController
 metadata:
   name: worker
 spec:
   replicas: 2
   selector:
     app: worker
   template:
     metadata:
       labels:
         app: worker
     spec:
       containers:
         - name: worker
           image: env/worker:v0.1
           ports:
           - containerPort: 3000
           env:
           - name: APP_DOMAIN
             value: "server"   # 传入service名字，后续脚本通过该名字从k8s获取到VIP
           - name: APP_PORT
             value: "3000"
 ➜  k8s  

部署验证

根据yaml文件创建rc/svc

 # create server-rc
 ➜  k8s  kubectl create -f server-rc.yml
 replicationcontroller/server created
 ➜  k8s  kubectl get rc
 NAME     DESIRED   CURRENT   READY   AGE
 server   1         1         1       15s

 # create service
 ➜  k8s  kubectl create -f server-svc.yml
 service/server created
 ➜  k8s
 ➜  ~  kubectl get svc
 NAME         TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)    AGE
 kubernetes   ClusterIP   10.96.0.1       <none>        443/TCP    24h
 server       ClusterIP   10.103.92.121   <none>        3000/TCP   38s
 ➜  ~  

 # create worker-rc
 ➜  k8s  kubectl create -f worker-rc.yml
 replicationcontroller/worker created
 ➜  k8s  kubectl get rc
 NAME     DESIRED   CURRENT   READY   AGE
 server   1         1         1       14m
 worker   2         2         2       10s
 ➜  k8s  kubectl get pods
 NAME           READY   STATUS    RESTARTS   AGE
 server-x67sw   1/1     Running   0          14m
 worker-45qjt   1/1     Running   0          15s
 worker-gvwlt   1/1     Running   0          15s
 ➜  k8s  

通过exec进入server/worker容器查看日志, 可以看到连接已经成功:

 ➜  k8s docker ps | grep python | awk '{print $1, $3 $4}'
 7b96a767637c "pythonworker.py"
 fb45f45be871 "pythonworker.py"
 e217df0ca52d "pythonserver.py"
 ➜  k8s docker exec -it e217df0ca52d /bin/bash
 root@server-x67sw:/env/server# ps -ef
 UID        PID  PPID  C STIME TTY          TIME CMD
 root         1     0  0 15:39 ?        00:00:00 python server.py
 root         7     0  0 15:40 pts/0    00:00:00 /bin/bash
 root        12     7  0 15:40 pts/0    00:00:00 ps -ef
 root@server-x67sw:/env/server# ls
 Dockerfile  log  server.py
 root@server-x67sw:/env/server# tail -f log/server.log
 2018-12-06 15:39:56 [INFO]  Waiting for connection...
 2018-12-06 15:54:06 [INFO]  ('10.32.0.9', 51266) connected
 2018-12-06 15:54:06 [INFO]  ('10.32.0.10', 47969) connected
 ^C
 root@server-x67sw:/env/server# exit
 exit
 ➜  k8s docker exec -it 7b96a767637c /bin/bash
 root@worker-gvwlt:/env/worker# ps -ef
 UID        PID  PPID  C STIME TTY          TIME CMD
 root         1     0  0 15:54 ?        00:00:00 python worker.py
 root         6     0  0 15:57 pts/0    00:00:00 /bin/bash
 root        11     6  0 15:57 pts/0    00:00:00 ps -ef
 root@worker-gvwlt:/env/worker# tail -f log/worker.log
 2018-12-06 15:56:31 [INFO]  [recv] hello, docker!
 ^C
 root@worker-gvwlt:/env/worker# exit
 exit
 ➜  k8s 

在宿主机直接使用VIP访问server成功(使用original/worker进行测试, ip地址通过配置文件读入)：

 ➜  worker  cat ini/config.ini | grep ip
 ip = 10.103.92.121
 ➜  worker  python worker.py
 [+][recv] hello, docker!
 [+][recv] hello, docker!
 ^C
 [-] signal(2) received, exit!
 ➜  worker